ROS index logo ROS Index
  1. Home
  2. Packages
  3. autoware_perception_launch
 
No version for distro humble showing github. Known supported distros are highlighted in the buttons above.
Package symbol

autoware_perception_launch package from autoware_launch repo

autoware_launch autoware_sample_designs autoware_control_config autoware_control_launch autoware_localization_config autoware_localization_launch autoware_map_config autoware_map_launch autoware_perception_config autoware_perception_launch autoware_planning_config autoware_planning_launch autoware_sensing_launch autoware_sensor_driver_launch awsim_labs_sensor_kit_description awsim_labs_sensor_kit_launch awsim_sensor_kit_description awsim_sensor_kit_launch carla_sensor_kit_description carla_sensor_kit_launch sample_sensor_kit_description sample_sensor_kit_launch single_lidar_sensor_kit_description single_lidar_sensor_kit_launch tier4_autoware_api_launch tier4_simulator_launch tier4_system_launch tier4_vehicle_launch awsim_labs_vehicle_description awsim_labs_vehicle_launch sample_vehicle_description sample_vehicle_launch

ROS Distro
github

API Docs
Browse Code

Package Summary

Version 0.53.0
License Apache License 2.0
Build type AMENT_CMAKE
Use RECOMMENDED

Repository Summary

Description
Checkout URI https://github.com/autowarefoundation/autoware_launch.git
VCS Type git
VCS Version main
Last Updated 2026-10-26
Dev Status UNKNOWN
Released UNRELEASED
Contributing Help Wanted (-)
Good First Issues (-)
Pull Requests to Review (-)

Package Description

The autoware_perception_launch package

Maintainers

  • Yukihiro Saito
  • Dai Nguyen
  • Taekjin Lee
  • Yoshi Ri
  • Masato Saeki

Authors

No additional authors.
autoware_universe_launch/autoware_perception_launch/README.md

autoware_perception_launch

Shared perception launch library. The system component launchers live in the product layer (autoware_launch/launch/components/component_perception.launch.xml for the lidar pipeline and component_traffic_light.launch.xml for traffic light recognition); this package hosts the perception feature launchers they combine:

launch/
├── common/                        # shared preprocess feature launchers (pointcloud downsample)
├── object_recognition/
│   ├── object_recognition_<mode>.launch.xml  # detection + tracking + prediction per sensor mode
│   ├── detection/
│   │   ├── detector/              # detectors
│   │   ├── filter/                # filters
│   │   └── merger/                # detection-side serial merger (used by the camera_lidar_radar_serial launcher)
│   └── prediction/
├── obstacle_segmentation/
├── occupancy_grid_map/
└── traffic_light_recognition/
design/module/                     # Autoware System Designer modules; same subtree as launch/
├── object_recognition/
├── obstacle_segmentation/
├── occupancy_grid_map/
└── traffic_light_recognition/

Launch files

  • object_recognition/object_recognition_<mode>.launch.xml — one feature launcher per sensor mode (camera, lidar, lidar_ptv3, lidar_radar, camera_lidar_radar, camera_lidar_radar_serial); hosts the detection stage inline together with the tracker and prediction, and pushes their module namespaces. Camera and radar input topics default to their absolute sensing paths (/sensing/camera/cameraN/..., /sensing/radar/detected_objects), so callers override only what differs. Most modes wire the multi-channel tracker to the detector outputs; camera_lidar_radar_serial instead merges the detectors in the detection stage (optionally regenerating detections through detection-by-tracker), tracks the single merged detected_objects channel, and joins radar tracked objects at the decorative tracker merger. camera runs camera-only detectors (BEV and camera VRU).
  • The clustering backend is file selection, not a flag: lidar runs rule-based euclidean clustering off the obstacle segmentation cloud, while lidar_ptv3 and lidar_radar cluster the PTv3 semantic-segmentation cloud and drop the pointcloud map filter and the object validator that only the rule-based pipeline uses. Both feed their cluster objects to the tracker on the lidar_clustering channel.
  • object_recognition/detection/{detector,filter}/ — single-purpose launchers wired together by the object-recognition launchers. merger/camera_lidar_merger.launch.xml chains the detection-side object mergers for the camera_lidar_radar_serial launcher.
  • obstacle_segmentation/ground_segmentation_<backend>.launch.xml — ground removal backends (rule_based, rule_based_cuda, ml); backend selection is file selection. ground_segmentation_{additional_lidars,ransac}.launch.xml fold extra lidars into the obstacle stream for kits that carry them.
  • occupancy_grid_map/, common/, traffic_light_recognition/ — feature launchers included directly by the system components.

Namespace and container addressing

Callers own the namespaces: the system component pushes perception and object_recognition; object_recognition_<mode>.launch.xml pushes detection/tracking/prediction. The object-recognition launchers address their internal topics through the fixed /perception/object_recognition namespace. Composable nodes load into the shared pointcloud container, whose name is exposed as pointcloud_container_name and aliased to the node/pointcloud_container argument of the detector launchers.

Config

Parameter files are resolved from a config/ tree addressed through the perception_config_pkg argument, which defaults to autoware_perception_config. A product can relocate the whole tree by setting perception_config_pkg once at its entry point — the launch configuration propagates through every include — but it must then ship the same config/ layout. Products that differ in only a few files instead override the corresponding *_param_path arguments, which every feature launcher declares with a default:

<include file="$(find-pkg-share autoware_perception_launch)/launch/object_recognition/object_recognition_lidar.launch.xml">
  <arg name="perception_config_pkg" value="my_perception_config"/>
  <arg name="FOO_NODE_param_path" value="..."/>
</include>

component_perception.launch.xml, component_traffic_light.launch.xml and planning_simulator/perception_simulation.launch.xml each declare perception_config_pkg, so a product switches the tree at the component it launches.

The launch/ tree is the interface downstream distributions adopt as-is; only package.xml, CMakeLists.txt, README.md, and changelogs stay package-local. Keep changes to this tree folder-merge compatible.

Package Dependencies

Please see <exec_depend> in package.xml.

CHANGELOG

Changelog for package autoware_perception_launch

0.53.0 (2026-10-13)

  • Merge remote-tracking branch 'origin/main' into tmp/bot/bump_version_base

  • feat(autoware_launch): run crop_box_filter standalone when agnocast is enabled (#1990)

    * feat(autoware_launch): run crop_box_filter standalone when agnocast is enabled autoware_universe#13384 moves CropBoxFilterComponent to agnocast_wrapper::Node on an agnocast-only executor, which the component containers do not set up. Start each of the eight crop_box_filter nodes as a standalone process when ENABLE_AGNOCAST=1 and keep loading them into their container otherwise. The node names, namespaces, remaps and parameters are the same in both branches.

    * refactor(autoware_launch): launch crop_box_filter through the autoware_node action ---------

  • feat(autoware_perception_launch): run scan_ground_filter standalone when agnocast is enabled (#1991)

    * feat(autoware_perception_launch): run scan_ground_filter standalone when agnocast is enabled autoware_universe#13385 moves ScanGroundFilterComponent to agnocast_wrapper::Node on an agnocast-only executor, which the component containers do not set up. Start each of the three scan_ground_filter nodes as a standalone process when ENABLE_AGNOCAST=1 and keep loading them into their container otherwise. The node names, namespaces, remaps and parameters are the same in both branches.

    * refactor(autoware_perception_launch): launch scan_ground_filter through the autoware_node action ---------

  • feat(autoware_launch): run voxel_grid_downsample_filter standalone when agnocast is enabled (#1992)

    • feat(autoware_launch): run voxel_grid_downsample_filter standalone when agnocast is enabled
    • style(pre-commit): autofix

    * refactor(autoware_launch): launch voxel_grid_downsample_filter through the autoware_node action ---------Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>

  • feat(autoware_sample_designs): compose the AD API, system and vehicle components from design modules (#1976)

    * fix(control, planning): connect the operation mode state and velocity limit inputs Wires collision_detector's operation_mode_state and remaining_distance_time_calculator's planning_velocity through the module designs; their node designs now declare the fixed topic names as remap targets.

    * fix(planning): connect the internal velocity limit path to the selector Forwards motion_velocity_planner's velocity_limit and clear_velocity_limit up to ScenarioPlanning and feeds them into external_velocity_limit_selector, whose internal inputs were unconnected.

    * fix(control, planning): connect the remaining module inputs to their sources Wires the operation mode transition manager's gate and selector inputs, the stop mode operator's vehicle status, the manual lane change handler's route availability and preferred primitive service, the parking costmap, and the route marker. Exposes the vehicle control mode service, the preferred lane service and the system hazard lights command as component interfaces.

    • fix(planning): follow the hazard lights selector input rename
    • fix(planning): add steering_status subscriber to various planning modules
    • feat(vehicle): add vehicle wrapper and launch file for vehicle interface
    • style(pre-commit): autofix
    • fix(planning): add planning factors publishers and connections for various modules
    • refactor(autoware_sample_designs): compose the AD API from design modules
    • Refactor ADAPI module structure and enhance planning factors
    • Updated AdapiRvizHelpers.module.yaml to include helper prefixes for instances and connections.
    • Modified AutowareSample.system.yaml to reorganize ADAPI components, introducing default_adapi for better structure.
    • Expanded planning factors in Planning.module.yaml, LaneDriving.module.yaml, ScenarioPlanning.module.yaml, and associated modules to include additional factors for improved behavior and motion planning.
    • Enhanced connections in behavior planning and motion planning modules to accommodate new planning factors.
    • Ensured consistency across scenario planning modules by aligning

File truncated at 100 lines see the full file

Package Dependencies

Deps Name
ament_cmake_auto
autoware_cmake
autoware_agnocast_wrapper
autoware_bevfusion
autoware_bytetrack
autoware_compare_map_segmentation
autoware_crosswalk_traffic_light_estimator
autoware_cuda_pointcloud_preprocessor
autoware_detected_object_feature_remover
autoware_detected_object_validation
autoware_detection_by_tracker
autoware_elevation_map_loader
autoware_euclidean_cluster
autoware_ground_segmentation
autoware_ground_segmentation_cuda
autoware_image_object_locator
autoware_image_projection_based_fusion
autoware_image_transport_decompressor
autoware_lidar_apollo_instance_segmentation
autoware_lidar_centerpoint
autoware_lidar_transfusion
autoware_map_based_prediction
autoware_multi_object_tracker
autoware_object_merger
autoware_object_sorter
autoware_occupancy_grid_map_outlier_filter
autoware_perception_config
autoware_pointcloud_preprocessor
autoware_probabilistic_occupancy_grid_map
autoware_ptv3
autoware_raindrop_cluster_filter
autoware_shape_estimation
autoware_simpl_prediction
autoware_simple_object_merger
autoware_tensorrt_bevdet
autoware_tensorrt_yolox
autoware_tracking_object_merger
autoware_traffic_light_arbiter
autoware_traffic_light_category_merger
autoware_traffic_light_classifier
autoware_traffic_light_fine_detector
autoware_traffic_light_map_based_detector
autoware_traffic_light_multi_camera_fusion
autoware_traffic_light_occlusion_predictor
autoware_traffic_light_selector
autoware_traffic_light_visualization
rclcpp_components
topic_tools
ament_lint_auto
autoware_lint_common

System Dependencies

No direct system dependencies.

Dependant Packages

Name Deps
autoware_launch
reaction_analyzer

Launch files

  • launch/common/pointcloud_downsample.launch.xml
      • pointcloud_container_name
      • cuda_pointcloud_preprocessing [default: false]
      • input/pointcloud
      • output/pointcloud
      • common_downsample_voxel_size_x [default: 0.05]
      • common_downsample_voxel_size_y [default: 0.05]
      • common_downsample_voxel_size_z [default: 0.05]
  • launch/object_recognition/detection/detector/camera_2d_detector.launch.xml
      • image_raw0 [default: /sensing/camera/camera0/image_raw]
      • image_raw1 [default: /sensing/camera/camera1/image_raw]
      • image_raw2 [default: /sensing/camera/camera2/image_raw]
      • image_raw3 [default: /sensing/camera/camera3/image_raw]
      • image_raw4 [default: /sensing/camera/camera4/image_raw]
      • image_raw5 [default: /sensing/camera/camera5/image_raw]
      • image_raw6 [default: /sensing/camera/camera6/image_raw]
      • image_raw7 [default: /sensing/camera/camera7/image_raw]
      • image_raw8 [default: /sensing/camera/camera8/image_raw]
      • image_raw9 [default: /sensing/camera/camera9/image_raw]
      • image_number [default: 1]
      • camera_index [default: 0]
      • use_bytetrack [default: true]
      • enable_visualizer [default: false]
      • camera_2d_detector/model_path
      • camera_2d_detector/label_path
      • camera_2d_detector/color_map_path
      • camera_2d_detector/roi_remap_path
      • camera_2d_detector/roi_to_semseg_remap_path
      • tensorrt_yolox_ns [default: ]
  • launch/object_recognition/detection/detector/camera_bev_detector.launch.xml
      • input/camera0/image
      • input/camera0/info
      • input/camera1/image
      • input/camera1/info
      • input/camera2/image
      • input/camera2/info
      • input/camera3/image
      • input/camera3/info
      • input/camera4/image
      • input/camera4/info
      • input/camera5/image
      • input/camera5/info
      • output/objects
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • bevdet_model_name [default: bevdet_one_lt_d]
      • bevdet_model_path [default: $(var data_path)/tensorrt_bevdet]
  • launch/object_recognition/detection/detector/camera_lidar_detector.launch.xml
      • ns
      • lidar_detection_model_type
      • lidar_detection_model_name
      • use_low_intensity_cluster_filter
      • use_image_segmentation_based_filter
      • number_of_cameras
      • input/camera0/image
      • input/camera0/info
      • input/camera0/rois
      • input/camera1/image
      • input/camera1/info
      • input/camera1/rois
      • input/camera2/image
      • input/camera2/info
      • input/camera2/rois
      • input/camera3/image
      • input/camera3/info
      • input/camera3/rois
      • input/camera4/image
      • input/camera4/info
      • input/camera4/rois
      • input/camera5/image
      • input/camera5/info
      • input/camera5/rois
      • input/camera6/image
      • input/camera6/info
      • input/camera6/rois
      • input/camera7/image
      • input/camera7/info
      • input/camera7/rois
      • input/camera8/image
      • input/camera8/info
      • input/camera8/rois
      • input/concatenation_info
      • segmentation_pointcloud_fusion_camera_ids
      • image_topic_name
      • sync_param_path
      • voxel_grid_based_euclidean_param_path
      • lidar_model_param_path
      • pointpainting_model_path [default: $(var data_path)/image_projection_based_fusion]
      • pointpainting_fusion_common_param_path
      • roi_cluster_fusion_param_path
      • node/pointcloud_container
      • input/pointcloud
      • input/pointcloud_map/pointcloud
      • input/obstacle_segmentation/pointcloud
      • output/ml_detector/objects
      • output/rule_detector/objects
      • camera_2d_detector/model_path
      • camera_2d_detector/label_path
      • camera_2d_detector/color_map_path
      • camera_2d_detector/roi_remap_path
      • camera_2d_detector/roi_to_semseg_remap_path
      • enable_2d_detection [default: false]
  • launch/object_recognition/detection/detector/camera_lidar_irregular_object_detector.launch.xml
      • pipeline_ns
      • node/pointcloud_container
      • output/objects [default: objects]
      • input/concatenation_info
      • input/pointcloud
      • irregular_object_crop_box_filter_node_param_path
      • irregular_object_scan_ground_filter_node_param_path
      • irregular_object_roi_pointcloud_fusion_node_param_path
      • irregular_object_fusion_camera_param_path
      • sync_param_path
  • launch/object_recognition/detection/detector/camera_vru_detector.launch.xml
      • input/camera0/info [default: /sensing/camera/camera0/camera_info]
      • input/camera0/rois [default: /perception/object_recognition/detection/rois0]
      • input/camera1/info [default: /sensing/camera/camera1/camera_info]
      • input/camera1/rois [default: /perception/object_recognition/detection/rois1]
      • input/camera2/info [default: /sensing/camera/camera2/camera_info]
      • input/camera2/rois [default: /perception/object_recognition/detection/rois2]
      • input/camera3/info [default: /sensing/camera/camera3/camera_info]
      • input/camera3/rois [default: /perception/object_recognition/detection/rois3]
      • input/camera4/info [default: /sensing/camera/camera4/camera_info]
      • input/camera4/rois [default: /perception/object_recognition/detection/rois4]
      • input/camera5/info [default: /sensing/camera/camera5/camera_info]
      • input/camera5/rois [default: /perception/object_recognition/detection/rois5]
      • input/camera6/info [default: /sensing/camera/camera6/camera_info]
      • input/camera6/rois [default: /perception/object_recognition/detection/rois6]
      • input/camera7/info [default: /sensing/camera/camera7/camera_info]
      • input/camera7/rois [default: /perception/object_recognition/detection/rois7]
      • output/objects [default: /perception/object_recognition/detection/camera_vru/objects]
      • bbox_object_locator_param_path
      • rois_ids [default: [0, 1]]
  • launch/object_recognition/detection/detector/irregular_object_detector.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • target_container
      • base_frame [default: base_link]
      • use_intra_process [default: true]
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • input/concatenation_info [default: /sensing/lidar/concatenated/pointcloud_info]
      • output/clusters
      • crop_box_filter_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/crop_box_filter.param.yaml]
      • scan_ground_filter_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/scan_ground_filter.param.yaml]
      • roi_pointcloud_fusion_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/roi_pointcloud_fusion.param.yaml]
      • fusion_camera_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/fusion_cameras.param.yaml]
      • sync_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/fusion_common.param.yaml]
  • launch/object_recognition/detection/detector/lidar_dnn_detector.launch.xml
      • lidar_detection_model_type
      • lidar_detection_model_name
      • lidar_model_param_path
      • bevfusion_model_path [default: $(var data_path)/bevfusion]
      • centerpoint_model_path [default: $(var data_path)/lidar_centerpoint]
      • centerpoint_common_param_path [default: $(find-pkg-share autoware_lidar_centerpoint)/config/centerpoint_common.param.yaml]
      • transfusion_model_path [default: $(var data_path)/lidar_transfusion]
      • use_short_range_detection [default: false]
      • lidar_short_range_detection_model_type
      • node/pointcloud_container
      • input/pointcloud
      • output/objects
      • output/short_range_objects
  • launch/object_recognition/detection/detector/lidar_rule_detector.launch.xml
      • ns
      • node/pointcloud_container
      • input/pointcloud_map/pointcloud
      • input/obstacle_segmentation/pointcloud
      • output/cluster_objects
      • output/objects
      • voxel_grid_based_euclidean_param_path
      • use_low_height_cropbox [default: false]
  • launch/object_recognition/detection/detector/lidar_semantic_segmentation_detector.launch.xml
      • euclidean_clustering/param_path
      • input/pointcloud
      • output/objects
  • launch/object_recognition/detection/detector/tracker_based_detector.launch.xml
      • detection_by_tracker_param_path
      • input/clusters
      • input/tracked_objects
      • output/objects
  • launch/object_recognition/detection/filter/object_filter.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • objects_filter_method [default: lanelet_filter]
      • object_recognition_detection_object_lanelet_filter_param_path [default: $(var perception_config_dir)/object_recognition/detection/object_filter/object_lanelet_filter.param.yaml]
      • object_recognition_detection_object_position_filter_param_path [default: $(var perception_config_dir)/object_recognition/detection/object_filter/object_position_filter.param.yaml]
      • input/objects
      • output/objects
  • launch/object_recognition/detection/filter/object_validator.launch.xml
      • objects_validation_method
      • object_recognition_detection_obstacle_pointcloud_based_validator_param_path
      • input/obstacle_pointcloud
      • input/objects
      • output/objects
  • launch/object_recognition/detection/filter/pointcloud_downsample_filter.launch.xml
      • target_container
      • use_intra_process [default: true]
      • input/pointcloud
      • output/pointcloud
      • downsample_voxel_size [default: 0.1]
  • launch/object_recognition/detection/filter/pointcloud_map_filter.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • target_container
      • use_intra_process [default: true]
      • input/pointcloud
      • output/pointcloud
      • output/downsampled/pointcloud [default: /perception/obstacle_segmentation/pointcloud_map_filtered/downsampled/pointcloud]
      • downsample_voxel_size [default: 0.1]
      • voxel_based_compare_map_filter_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/pointcloud_filter/voxel_based_compare_map_filter.param.yaml]
  • launch/object_recognition/detection/filter/radar_filter.launch.xml
      • object_sorter_param_path
      • radar_lanelet_filtering_range_param_path
      • input/radar
      • output/objects
  • launch/object_recognition/detection/merger/camera_lidar_merger.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • object_recognition_detection_object_merger_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/detection/object_merger/data_association_matrix.param.yaml]
      • object_recognition_detection_object_merger_distance_threshold_list_path [default: $(var perception_config_dir)/object_recognition/detection/object_merger/overlapped_judge.param.yaml]
      • ml_camera_lidar_object_association_merger_param_path [default: $(var perception_config_dir)/object_recognition/detection/object_merger/ml_camera_lidar_object_association_merger.param.yaml]
      • roi_detected_object_fusion_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/roi_detected_object_fusion.param.yaml]
      • sync_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/fusion_common.param.yaml]
      • lidar_detection_model_type
      • use_detection_by_tracker
      • use_irregular_object_detector
      • use_roi_detected_object_validator
      • euclidean_clustering/shape_policy
      • use_object_filter
      • objects_filter_method
      • number_of_cameras
      • input/camera0/image
      • input/camera0/info
      • input/camera0/rois
      • input/camera1/image
      • input/camera1/info
      • input/camera1/rois
      • input/camera2/image
      • input/camera2/info
      • input/camera2/rois
      • input/camera3/image
      • input/camera3/info
      • input/camera3/rois
      • input/camera4/image
      • input/camera4/info
      • input/camera4/rois
      • input/camera5/image
      • input/camera5/info
      • input/camera5/rois
      • input/camera6/image
      • input/camera6/info
      • input/camera6/rois
      • input/camera7/image
      • input/camera7/info
      • input/camera7/rois
      • input/camera8/image
      • input/camera8/info
      • input/camera8/rois
      • input/concatenation_info [default: /sensing/lidar/concatenated/pointcloud_info]
      • input/lidar_ml/objects
      • input/lidar_cluster/objects
      • input/irregular_object/objects [default: ]
      • input/detection_by_tracker/objects
      • output/objects [default: objects]
      • alpha_merger_priority_mode [default: 0]
  • launch/object_recognition/object_recognition_camera.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • use_camera_vru_detector [default: false]
      • camera_vru_detector_rois_ids [default: [0]]
      • use_camera_bev_detector [default: true]
      • use_vector_map [default: true]
      • tracker_publish_merged_objects [default: false]
      • input/camera0/image [default: /sensing/camera/camera0/image_rect_color]
      • input/camera0/info [default: /sensing/camera/camera0/camera_info]
      • input/camera0/rois [default: /perception/object_recognition/detection/rois0]
      • input/camera1/image [default: /sensing/camera/camera1/image_rect_color]
      • input/camera1/info [default: /sensing/camera/camera1/camera_info]
      • input/camera1/rois [default: /perception/object_recognition/detection/rois1]
      • input/camera2/image [default: /sensing/camera/camera2/image_rect_color]
      • input/camera2/info [default: /sensing/camera/camera2/camera_info]
      • input/camera2/rois [default: /perception/object_recognition/detection/rois2]
      • input/camera3/image [default: /sensing/camera/camera3/image_rect_color]
      • input/camera3/info [default: /sensing/camera/camera3/camera_info]
      • input/camera3/rois [default: /perception/object_recognition/detection/rois3]
      • input/camera4/image [default: /sensing/camera/camera4/image_rect_color]
      • input/camera4/info [default: /sensing/camera/camera4/camera_info]
      • input/camera4/rois [default: /perception/object_recognition/detection/rois4]
      • input/camera5/image [default: /sensing/camera/camera5/image_rect_color]
      • input/camera5/info [default: /sensing/camera/camera5/camera_info]
      • input/camera5/rois [default: /perception/object_recognition/detection/rois5]
      • input/camera6/image [default: /sensing/camera/camera6/image_rect_color]
      • input/camera6/info [default: /sensing/camera/camera6/camera_info]
      • input/camera6/rois [default: /perception/object_recognition/detection/rois6]
      • input/camera7/image [default: /sensing/camera/camera7/image_rect_color]
      • input/camera7/info [default: /sensing/camera/camera7/camera_info]
      • input/camera7/rois [default: /perception/object_recognition/detection/rois7]
      • output/tracker_merged_objects [default: /perception/object_recognition/detection/objects]
      • output/tracked_objects [default: /perception/object_recognition/tracking/objects]
      • output/predicted_objects [default: /perception/object_recognition/objects]
      • near_range_camera_vru_param_path [default: $(var perception_config_dir)/object_recognition/detection/camera_vru_detection/near_range_camera_vru_detector.param.yaml]
      • object_recognition_tracking_multi_object_tracker_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/data_association_matrix.param.yaml]
      • object_recognition_tracking_multi_object_tracker_input_channels_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/input_channels.param.yaml]
      • object_recognition_tracking_multi_object_tracker_node_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/multi_object_tracker_node.param.yaml]
  • launch/object_recognition/object_recognition_camera_lidar_radar.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • lidar_detection_model [default: centerpoint]
      • use_short_range_detection [default: false]
      • lidar_short_range_detection_model_type [default: centerpoint_short_range]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • bevfusion_model_path [default: $(var data_path)/bevfusion]
      • pointpainting_model_path [default: $(var data_path)/image_projection_based_fusion]
      • camera_2d_detector/model_path [default: $(var data_path)/tensorrt_yolox/yolox-sPlus-opt-pseudoV2-T4-960x960-T4-seg16cls.onnx]
      • camera_2d_detector/label_path [default: $(var data_path)/tensorrt_yolox/label.txt]
      • camera_2d_detector/color_map_path [default: $(var data_path)/tensorrt_yolox/semseg_color_map.csv]
      • camera_2d_detector/roi_remap_path [default: $(var perception_config_dir)/traffic_light_recognition/tensorrt_yolox/roi_label_remap.csv]
      • camera_2d_detector/roi_to_semseg_remap_path [default: $(var perception_config_dir)/traffic_light_recognition/tensorrt_yolox/roi_to_semseg_label_remap.csv]
      • use_semantic_segmentation_ptv3 [default: false]
      • use_object_filter [default: true]
      • use_pointcloud_map [default: true]
      • use_validator [default: false]
      • objects_validation_method [default: obstacle_pointcloud]
      • use_low_height_cropbox [default: false]
      • use_low_intensity_cluster_filter [default: true]
      • use_image_segmentation_based_filter [default: false]
      • use_irregular_object_detector [default: true]
      • use_camera_vru_detector [default: false]
      • camera_vru_detector_rois_ids [default: [0]]
      • use_camera_only_detection [default: false]
      • use_vector_map [default: true]
      • tracker_publish_merged_objects [default: false]
      • number_of_cameras [default: 6]
      • pointcloud_container_name [default: pointcloud_container]
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • input/semantic_segmentation/pointcloud [default: /perception/segmented/pointcloud]
      • input/obstacle_segmentation/pointcloud [default: /perception/obstacle_segmentation/pointcloud]
      • input/camera0/image [default: /sensing/camera/camera0/image_rect_color]
      • input/camera0/info [default: /sensing/camera/camera0/camera_info]
      • input/camera0/rois [default: /perception/object_recognition/detection/rois0]
      • input/camera1/image [default: /sensing/camera/camera1/image_rect_color]
      • input/camera1/info [default: /sensing/camera/camera1/camera_info]
      • input/camera1/rois [default: /perception/object_recognition/detection/rois1]
      • input/camera2/image [default: /sensing/camera/camera2/image_rect_color]
      • input/camera2/info [default: /sensing/camera/camera2/camera_info]
      • input/camera2/rois [default: /perception/object_recognition/detection/rois2]
      • input/camera3/image [default: /sensing/camera/camera3/image_rect_color]
      • input/camera3/info [default: /sensing/camera/camera3/camera_info]
      • input/camera3/rois [default: /perception/object_recognition/detection/rois3]
      • input/camera4/image [default: /sensing/camera/camera4/image_rect_color]
      • input/camera4/info [default: /sensing/camera/camera4/camera_info]
      • input/camera4/rois [default: /perception/object_recognition/detection/rois4]
      • input/camera5/image [default: /sensing/camera/camera5/image_rect_color]
      • input/camera5/info [default: /sensing/camera/camera5/camera_info]
      • input/camera5/rois [default: /perception/object_recognition/detection/rois5]
      • input/camera6/image [default: /sensing/camera/camera6/image_rect_color]
      • input/camera6/info [default: /sensing/camera/camera6/camera_info]
      • input/camera6/rois [default: /perception/object_recognition/detection/rois6]
      • input/camera7/image [default: /sensing/camera/camera7/image_rect_color]
      • input/camera7/info [default: /sensing/camera/camera7/camera_info]
      • input/camera7/rois [default: /perception/object_recognition/detection/rois7]
      • input/camera8/image [default: /sensing/camera/camera8/image_rect_color]
      • input/camera8/info [default: /sensing/camera/camera8/camera_info]
      • input/camera8/rois [default: /perception/object_recognition/detection/rois8]
      • input/concatenation_info [default: /sensing/lidar/concatenated/pointcloud_info]
      • input/radar [default: /sensing/radar/detected_objects]
      • input/camera_only/objects [default: /perception/object_recognition/detection/camera_only/objects]
      • input/camera_only/channel [default: camera_streampetr]
      • image_topic_name [default: image_rect_color]
      • segmentation_pointcloud_fusion_camera_ids [default: [0,2,4]]
      • output/tracker_merged_objects [default: /perception/object_recognition/detection/objects]
      • output/tracked_objects [default: /perception/object_recognition/tracking/objects]
      • output/predicted_objects [default: /perception/object_recognition/objects]
      • sync_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/fusion_common.param.yaml]
      • roi_cluster_fusion_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/roi_cluster_fusion.param.yaml]
      • pointpainting_fusion_common_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/pointpainting_common.param.yaml]
      • irregular_object_crop_box_filter_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/crop_box_filter.param.yaml]
      • irregular_object_scan_ground_filter_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/scan_ground_filter.param.yaml]
      • irregular_object_roi_pointcloud_fusion_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/roi_pointcloud_fusion.param.yaml]
      • irregular_object_fusion_camera_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/fusion_cameras.param.yaml]
      • lidar_model_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model]
      • centerpoint_common_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model/centerpoint_common.param.yaml]
      • voxel_grid_based_euclidean_param_path [default: $(var perception_config_dir)/object_recognition/detection/clustering/voxel_grid_based_euclidean_cluster.param.yaml]
      • radar_lanelet_filtering_range_param_path [default: $(var perception_config_dir)/object_recognition/detection/object_filter/radar_lanelet_filter.param.yaml]
      • object_recognition_detection_object_sorter_radar_param_path [default: $(var perception_config_dir)/object_recognition/detection/radar_filter/detected_object_sorter.param.yaml]
      • near_range_camera_vru_param_path [default: $(var perception_config_dir)/object_recognition/detection/camera_vru_detection/near_range_camera_vru_detector.param.yaml]
      • object_recognition_detection_obstacle_pointcloud_based_validator_param_path [default: $(var perception_config_dir)/object_recognition/detection/detected_object_validation/obstacle_pointcloud_based_validator.param.yaml]
      • object_recognition_tracking_multi_object_tracker_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/data_association_matrix.param.yaml]
      • object_recognition_tracking_multi_object_tracker_input_channels_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/input_channels.param.yaml]
      • object_recognition_tracking_multi_object_tracker_node_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/multi_object_tracker_node.param.yaml]
      • euclidean_clustering/param_path [default: $(var perception_config_dir)/object_recognition/detection/euclidean_cluster/label_based_euclidean_cluster.param.yaml]
  • launch/object_recognition/object_recognition_camera_lidar_radar_serial.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • lidar_detection_model [default: centerpoint]
      • use_short_range_detection [default: false]
      • lidar_short_range_detection_model_type [default: centerpoint_short_range]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • bevfusion_model_path [default: $(var data_path)/bevfusion]
      • pointpainting_model_path [default: $(var data_path)/image_projection_based_fusion]
      • camera_2d_detector/model_path [default: $(var data_path)/tensorrt_yolox/yolox-sPlus-opt-pseudoV2-T4-960x960-T4-seg16cls.onnx]
      • camera_2d_detector/label_path [default: $(var data_path)/tensorrt_yolox/label.txt]
      • camera_2d_detector/color_map_path [default: $(var data_path)/tensorrt_yolox/semseg_color_map.csv]
      • camera_2d_detector/roi_remap_path [default: $(var perception_config_dir)/traffic_light_recognition/tensorrt_yolox/roi_label_remap.csv]
      • camera_2d_detector/roi_to_semseg_remap_path [default: $(var perception_config_dir)/traffic_light_recognition/tensorrt_yolox/roi_to_semseg_label_remap.csv]
      • use_object_filter [default: true]
      • objects_filter_method [default: lanelet_filter]
      • use_pointcloud_map [default: true]
      • use_validator [default: false]
      • objects_validation_method [default: obstacle_pointcloud]
      • use_roi_detected_object_validator [default: true]
      • use_detection_by_tracker [default: true]
      • use_low_height_cropbox [default: false]
      • use_low_intensity_cluster_filter [default: true]
      • use_image_segmentation_based_filter [default: false]
      • use_irregular_object_detector [default: true]
      • ml_camera_lidar_merger_priority_mode [default: 0]
      • use_vector_map [default: true]
      • number_of_cameras [default: 6]
      • pointcloud_container_name [default: pointcloud_container]
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • input/obstacle_segmentation/pointcloud [default: /perception/obstacle_segmentation/pointcloud]
      • input/camera0/image [default: /sensing/camera/camera0/image_rect_color]
      • input/camera0/info [default: /sensing/camera/camera0/camera_info]
      • input/camera0/rois [default: /perception/object_recognition/detection/rois0]
      • input/camera1/image [default: /sensing/camera/camera1/image_rect_color]
      • input/camera1/info [default: /sensing/camera/camera1/camera_info]
      • input/camera1/rois [default: /perception/object_recognition/detection/rois1]
      • input/camera2/image [default: /sensing/camera/camera2/image_rect_color]
      • input/camera2/info [default: /sensing/camera/camera2/camera_info]
      • input/camera2/rois [default: /perception/object_recognition/detection/rois2]
      • input/camera3/image [default: /sensing/camera/camera3/image_rect_color]
      • input/camera3/info [default: /sensing/camera/camera3/camera_info]
      • input/camera3/rois [default: /perception/object_recognition/detection/rois3]
      • input/camera4/image [default: /sensing/camera/camera4/image_rect_color]
      • input/camera4/info [default: /sensing/camera/camera4/camera_info]
      • input/camera4/rois [default: /perception/object_recognition/detection/rois4]
      • input/camera5/image [default: /sensing/camera/camera5/image_rect_color]
      • input/camera5/info [default: /sensing/camera/camera5/camera_info]
      • input/camera5/rois [default: /perception/object_recognition/detection/rois5]
      • input/camera6/image [default: /sensing/camera/camera6/image_rect_color]
      • input/camera6/info [default: /sensing/camera/camera6/camera_info]
      • input/camera6/rois [default: /perception/object_recognition/detection/rois6]
      • input/camera7/image [default: /sensing/camera/camera7/image_rect_color]
      • input/camera7/info [default: /sensing/camera/camera7/camera_info]
      • input/camera7/rois [default: /perception/object_recognition/detection/rois7]
      • input/camera8/image [default: /sensing/camera/camera8/image_rect_color]
      • input/camera8/info [default: /sensing/camera/camera8/camera_info]
      • input/camera8/rois [default: /perception/object_recognition/detection/rois8]
      • input/concatenation_info [default: /sensing/lidar/concatenated/pointcloud_info]
      • input/radar/tracked_objects [default: /sensing/radar/tracked_objects]
      • image_topic_name [default: image_rect_color]
      • segmentation_pointcloud_fusion_camera_ids [default: [0,2,4]]
      • output/tracked_objects [default: /perception/object_recognition/tracking/objects]
      • output/predicted_objects [default: /perception/object_recognition/objects]
      • sync_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/fusion_common.param.yaml]
      • roi_cluster_fusion_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/roi_cluster_fusion.param.yaml]
      • pointpainting_fusion_common_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/pointpainting_common.param.yaml]
      • irregular_object_crop_box_filter_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/crop_box_filter.param.yaml]
      • irregular_object_scan_ground_filter_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/scan_ground_filter.param.yaml]
      • irregular_object_roi_pointcloud_fusion_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/roi_pointcloud_fusion.param.yaml]
      • irregular_object_fusion_camera_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/fusion_cameras.param.yaml]
      • lidar_model_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model]
      • centerpoint_common_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model/centerpoint_common.param.yaml]
      • voxel_grid_based_euclidean_param_path [default: $(var perception_config_dir)/object_recognition/detection/clustering/voxel_grid_based_euclidean_cluster.param.yaml]
      • detection_by_tracker_param_path [default: $(var perception_config_dir)/object_recognition/detection/detection_by_tracker/detection_by_tracker.param.yaml]
      • object_recognition_detection_obstacle_pointcloud_based_validator_param_path [default: $(var perception_config_dir)/object_recognition/detection/detected_object_validation/obstacle_pointcloud_based_validator.param.yaml]
      • object_recognition_detection_object_merger_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/detection/object_merger/data_association_matrix.param.yaml]
      • object_recognition_detection_object_merger_distance_threshold_list_path [default: $(var perception_config_dir)/object_recognition/detection/object_merger/overlapped_judge.param.yaml]
      • ml_camera_lidar_object_association_merger_param_path [default: $(var perception_config_dir)/object_recognition/detection/object_merger/ml_camera_lidar_object_association_merger.param.yaml]
      • roi_detected_object_fusion_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/roi_detected_object_fusion.param.yaml]
      • object_recognition_tracking_multi_object_tracker_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/data_association_matrix.param.yaml]
      • object_recognition_tracking_multi_object_tracker_input_channels_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/input_channels.param.yaml]
      • object_recognition_tracking_multi_object_tracker_node_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/multi_object_tracker_node.param.yaml]
      • object_recognition_tracking_radar_tracked_object_sorter_param_path [default: $(var perception_config_dir)/object_recognition/tracking/radar_object_tracker/tracked_object_sorter.param.yaml]
      • object_recognition_tracking_radar_tracked_object_lanelet_filter_param_path [default: $(var perception_config_dir)/object_recognition/tracking/radar_object_tracker/tracked_object_lanelet_filter.param.yaml]
      • object_recognition_tracking_object_merger_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/tracking/tracking_object_merger/data_association_matrix.param.yaml]
      • object_recognition_tracking_object_merger_node_param_path [default: $(var perception_config_dir)/object_recognition/tracking/tracking_object_merger/decorative_tracker_merger.param.yaml]
      • euclidean_clustering/shape_policy [default: 0]
  • launch/object_recognition/object_recognition_lidar.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • lidar_detection_model [default: centerpoint]
      • use_short_range_detection [default: false]
      • lidar_short_range_detection_model_type [default: centerpoint_short_range]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • bevfusion_model_path [default: $(var data_path)/bevfusion]
      • use_pointcloud_map [default: true]
      • use_validator [default: false]
      • objects_validation_method [default: obstacle_pointcloud]
      • use_object_filter [default: true]
      • objects_filter_method [default: lanelet_filter]
      • use_low_height_cropbox [default: false]
      • use_vector_map [default: true]
      • tracker_publish_merged_objects [default: false]
      • pointcloud_container_name [default: pointcloud_container]
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • input/obstacle_segmentation/pointcloud [default: /perception/obstacle_segmentation/pointcloud]
      • output/tracker_merged_objects [default: /perception/object_recognition/detection/objects]
      • output/tracked_objects [default: /perception/object_recognition/tracking/objects]
      • output/predicted_objects [default: /perception/object_recognition/objects]
      • lidar_model_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model]
      • centerpoint_common_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model/centerpoint_common.param.yaml]
      • voxel_grid_based_euclidean_param_path [default: $(var perception_config_dir)/object_recognition/detection/clustering/voxel_grid_based_euclidean_cluster.param.yaml]
      • object_recognition_detection_obstacle_pointcloud_based_validator_param_path [default: $(var perception_config_dir)/object_recognition/detection/detected_object_validation/obstacle_pointcloud_based_validator.param.yaml]
      • object_recognition_tracking_multi_object_tracker_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/data_association_matrix.param.yaml]
      • object_recognition_tracking_multi_object_tracker_input_channels_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/input_channels.param.yaml]
      • object_recognition_tracking_multi_object_tracker_node_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/multi_object_tracker_node.param.yaml]
  • launch/object_recognition/object_recognition_lidar_ptv3.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • lidar_detection_model [default: centerpoint]
      • use_short_range_detection [default: false]
      • lidar_short_range_detection_model_type [default: centerpoint_short_range]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • bevfusion_model_path [default: $(var data_path)/bevfusion]
      • use_vector_map [default: true]
      • tracker_publish_merged_objects [default: false]
      • pointcloud_container_name [default: pointcloud_container]
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • input/semantic_segmentation/pointcloud [default: /perception/segmented/pointcloud]
      • output/tracker_merged_objects [default: /perception/object_recognition/detection/objects]
      • output/tracked_objects [default: /perception/object_recognition/tracking/objects]
      • output/predicted_objects [default: /perception/object_recognition/objects]
      • lidar_model_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model]
      • centerpoint_common_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model/centerpoint_common.param.yaml]
      • object_recognition_tracking_multi_object_tracker_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/data_association_matrix.param.yaml]
      • object_recognition_tracking_multi_object_tracker_input_channels_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/input_channels.param.yaml]
      • object_recognition_tracking_multi_object_tracker_node_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/multi_object_tracker_node.param.yaml]
      • euclidean_clustering/param_path [default: $(var perception_config_dir)/object_recognition/detection/euclidean_cluster/label_based_euclidean_cluster.param.yaml]
  • launch/object_recognition/object_recognition_lidar_radar.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • lidar_detection_model [default: centerpoint]
      • use_short_range_detection [default: false]
      • lidar_short_range_detection_model_type [default: centerpoint_short_range]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • bevfusion_model_path [default: $(var data_path)/bevfusion]
      • use_low_height_cropbox [default: false]
      • use_vector_map [default: true]
      • tracker_publish_merged_objects [default: false]
      • pointcloud_container_name [default: pointcloud_container]
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • input/semantic_segmentation/pointcloud [default: /perception/segmented/pointcloud]
      • input/radar [default: /sensing/radar/detected_objects]
      • output/tracker_merged_objects [default: /perception/object_recognition/detection/objects]
      • output/tracked_objects [default: /perception/object_recognition/tracking/objects]
      • output/predicted_objects [default: /perception/object_recognition/objects]
      • lidar_model_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model]
      • centerpoint_common_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model/centerpoint_common.param.yaml]
      • voxel_grid_based_euclidean_param_path [default: $(var perception_config_dir)/object_recognition/detection/clustering/voxel_grid_based_euclidean_cluster.param.yaml]
      • radar_lanelet_filtering_range_param_path [default: $(var perception_config_dir)/object_recognition/detection/object_filter/radar_lanelet_filter.param.yaml]
      • object_recognition_detection_object_sorter_radar_param_path [default: $(var perception_config_dir)/object_recognition/detection/radar_filter/detected_object_sorter.param.yaml]
      • object_recognition_tracking_multi_object_tracker_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/data_association_matrix.param.yaml]
      • object_recognition_tracking_multi_object_tracker_input_channels_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/input_channels.param.yaml]
      • object_recognition_tracking_multi_object_tracker_node_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/multi_object_tracker_node.param.yaml]
      • euclidean_clustering/param_path [default: $(var perception_config_dir)/object_recognition/detection/euclidean_cluster/label_based_euclidean_cluster.param.yaml]
  • launch/object_recognition/prediction/prediction.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • use_vector_map [default: false]
      • prediction_model_type [default: map_based]
      • input/objects [default: /perception/object_recognition/tracking/objects]
      • output/objects [default: /perception/object_recognition/objects]
      • object_recognition_prediction_map_based_prediction_param_path [default: $(var perception_config_dir)/object_recognition/prediction/map_based_prediction.param.yaml]
  • launch/obstacle_segmentation/ground_segmentation_additional_lidars.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • target_container
      • lidar_name
      • base_frame [default: base_link]
      • use_intra_process [default: true]
      • input/main/pointcloud
      • input/lidar/pointcloud [default: /sensing/lidar/$(var lidar_name)/pointcloud]
      • output/pointcloud
      • output/pointcloud_info [default: $(var output/pointcloud)_info]
      • crop_box_margin_max_z [default: 2.5]
      • crop_box_margin_min_z [default: -2.5]
      • crop_box_filter_node_param_path [default: $(var perception_config_dir)/obstacle_segmentation/ground_segmentation/crop_box_filter.param.yaml]
      • scan_ground_filter_node_param_path [default: $(var perception_config_dir)/obstacle_segmentation/ground_segmentation/scan_ground_filter.param.yaml]
  • launch/obstacle_segmentation/ground_segmentation_ml.launch.xml
      • pointcloud_container_name [default: pointcloud_container]
      • use_pointcloud_container [default: true]
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • output/pointcloud [default: /perception/obstacle_segmentation/pointcloud]
      • output/pointcloud/segmentation [default: /perception/segmented/pointcloud]
      • output/pointcloud/visualization [default: /perception/segmented/pointcloud/visualization]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • model_variant [default: ptv3]
      • model_path [default: $(var data_path)/ptv3]
  • launch/obstacle_segmentation/ground_segmentation_ransac.launch.xml
      • target_container
      • base_frame [default: base_link]
      • use_intra_process [default: true]
      • input_topics
      • input/main/pointcloud
      • output/pointcloud
      • output/pointcloud_info [default: $(var output/pointcloud)_info]
      • detection_area_crop_box_filter_node_param_path
      • ransac_ground_filter_node_param_path [default: $(find-pkg-share autoware_ground_segmentation)/config/ransac_ground_filter.param.yaml]
  • launch/obstacle_segmentation/ground_segmentation_rule_based.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • target_container
      • base_frame [default: base_link]
      • use_intra_process [default: true]
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • output/pointcloud [default: /perception/obstacle_segmentation/pointcloud]
      • crop_box_margin_max_z [default: 2.5]
      • crop_box_margin_min_z [default: -2.5]
      • crop_box_filter_node_param_path [default: $(var perception_config_dir)/obstacle_segmentation/ground_segmentation/crop_box_filter.param.yaml]
      • scan_ground_filter_node_param_path [default: $(var perception_config_dir)/obstacle_segmentation/ground_segmentation/scan_ground_filter.param.yaml]
  • launch/obstacle_segmentation/ground_segmentation_rule_based_cuda.launch.xml
      • target_container
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • output/pointcloud [default: /perception/obstacle_segmentation/pointcloud]
      • output/ground_pointcloud [default: /perception/obstacle_segmentation/ground_pointcloud]
      • cuda_scan_ground_segmentation_filter_node_param_path [default: $(find-pkg-share autoware_ground_segmentation_cuda)/config/cuda_scan_ground_segmentation_filter.param.yaml]
  • launch/occupancy_grid_map/probabilistic_occupancy_grid_map.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • input/obstacle_pointcloud [default: concatenated/pointcloud]
      • input/raw_pointcloud [default: no_ground/oneshot/pointcloud]
      • output [default: /perception/occupancy_grid_map/map]
      • use_intra_process [default: false]
      • use_multithread [default: false]
      • pointcloud_container_name [default: pointcloud_container]
      • occupancy_grid_map_method
      • occupancy_grid_map_param_path [default: $(var perception_config_dir)/occupancy_grid_map/$(var occupancy_grid_map_method).param.yaml]
      • occupancy_grid_map_updater
      • occupancy_grid_map_updater_param_path [default: $(var perception_config_dir)/occupancy_grid_map/$(var occupancy_grid_map_updater)_updater.param.yaml]
      • input_obstacle_pointcloud [default: false]
      • input_obstacle_and_raw_pointcloud [default: true]
      • use_pointcloud_container [default: true]
  • launch/traffic_light_recognition/traffic_light_detector_fine.launch.xml
    • Fine-detection chain for one traffic light camera: map based detector proposes rough ROIs, the fine detector refines them, classifiers label them.
      • perception_config_pkg [default: autoware_perception_config]
      • camera_namespace
      • enable_image_decompressor [default: true]
      • input/vector_map [default: /map/vector_map]
      • input/route [default: /planning/mission_planning/route]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • map_based_detector_param_path [default: $(find-pkg-share autoware_traffic_light_map_based_detector)/config/traffic_light_map_based_detector.param.yaml]
      • fine_detector_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_fine_detector/traffic_light_fine_detector.param.yaml]
      • car_classifier_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_classifier/car_traffic_light_classifier.param.yaml]
      • pedestrian_classifier_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_classifier/pedestrian_traffic_light_classifier.param.yaml]
      • lamp_recognizer_ml_param_path [default: $(var data_path)/traffic_light_classifier/lamp_recognizer_ml.param.yaml]
      • roi_visualizer_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_visualization/traffic_light_roi_visualizer.param.yaml]
      • fine_detector_model_path [default: $(var data_path)/traffic_light_fine_detector/tlr_car_ped_yolox_s_batch_6.onnx]
      • fine_detector_label_path [default: $(var data_path)/traffic_light_fine_detector/tlr_labels.txt]
      • car_classifier_model_path [default: $(var data_path)/traffic_light_classifier/traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • car_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels.txt]
      • pedestrian_classifier_model_path [default: $(var data_path)/traffic_light_classifier/ped_traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • pedestrian_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels_ped.txt]
      • car_classifier_type [default: 1]
      • pedestrian_classifier_type [default: 1]
  • launch/traffic_light_recognition/traffic_light_detector_rough.launch.xml
    • Rough-detection chain for one traffic light camera: map based detector ROIs feed the classifiers directly, without a high accuracy detector.
      • perception_config_pkg [default: autoware_perception_config]
      • camera_namespace
      • enable_image_decompressor [default: true]
      • input/vector_map [default: /map/vector_map]
      • input/route [default: /planning/mission_planning/route]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • map_based_detector_param_path [default: $(find-pkg-share autoware_traffic_light_map_based_detector)/config/traffic_light_map_based_detector.param.yaml]
      • car_classifier_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_classifier/car_traffic_light_classifier.param.yaml]
      • pedestrian_classifier_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_classifier/pedestrian_traffic_light_classifier.param.yaml]
      • lamp_recognizer_ml_param_path [default: $(var data_path)/traffic_light_classifier/lamp_recognizer_ml.param.yaml]
      • roi_visualizer_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_visualization/traffic_light_roi_visualizer.param.yaml]
      • car_classifier_model_path [default: $(var data_path)/traffic_light_classifier/traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • car_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels.txt]
      • pedestrian_classifier_model_path [default: $(var data_path)/traffic_light_classifier/ped_traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • pedestrian_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels_ped.txt]
      • car_classifier_type [default: 1]
      • pedestrian_classifier_type [default: 1]
  • launch/traffic_light_recognition/traffic_light_detector_whole_image.launch.xml
    • Whole-image detection chain for one traffic light camera: yolox detects over the full image, the selector matches detections to map based ROIs, classifiers label them.
      • perception_config_pkg [default: autoware_perception_config]
      • camera_namespace
      • enable_image_decompressor [default: true]
      • input/vector_map [default: /map/vector_map]
      • input/route [default: /planning/mission_planning/route]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • map_based_detector_param_path [default: $(find-pkg-share autoware_traffic_light_map_based_detector)/config/traffic_light_map_based_detector.param.yaml]
      • whole_image_detector_param_path [default: $(var perception_config_dir)/traffic_light_recognition/tensorrt_yolox/yolox_traffic_light_detector.param.yaml]
      • car_classifier_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_classifier/car_traffic_light_classifier.param.yaml]
      • pedestrian_classifier_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_classifier/pedestrian_traffic_light_classifier.param.yaml]
      • lamp_recognizer_ml_param_path [default: $(var data_path)/traffic_light_classifier/lamp_recognizer_ml.param.yaml]
      • roi_visualizer_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_visualization/traffic_light_roi_visualizer.param.yaml]
      • whole_image_detector_model_path [default: $(var data_path)/tensorrt_yolox/yolox_s_car_ped_tl_detector_960_960_batch_1.onnx]
      • whole_image_detector_label_path [default: $(var data_path)/tensorrt_yolox/car_ped_tl_detector_labels.txt]
      • car_classifier_model_path [default: $(var data_path)/traffic_light_classifier/traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • car_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels.txt]
      • pedestrian_classifier_model_path [default: $(var data_path)/traffic_light_classifier/ped_traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • pedestrian_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels_ped.txt]
      • car_classifier_type [default: 1]
      • pedestrian_classifier_type [default: 1]
  • launch/traffic_light_recognition/traffic_light_fusion.launch.xml
    • Post-process shared by all traffic light cameras: multi camera fusion, V2X arbitration, pedestrian signal estimation and map visualization.
      • perception_config_pkg [default: autoware_perception_config]
      • camera_namespaces [default: [camera6, camera7]]
      • input/vector_map [default: /map/vector_map]
      • external/traffic_signals [default: /perception/traffic_light_recognition/external/traffic_signals]
      • internal/traffic_signals [default: /perception/traffic_light_recognition/internal/traffic_signals]
      • judged/traffic_signals [default: /perception/traffic_light_recognition/judged/traffic_signals]
      • output/traffic_signals [default: /perception/traffic_light_recognition/traffic_signals]
      • multi_camera_fusion_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_multi_camera_fusion/traffic_light_multi_camera_fusion.param.yaml]
      • arbiter_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_arbiter/traffic_light_arbiter.param.yaml]
      • crosswalk_estimator_param_path [default: $(var perception_config_dir)/traffic_light_recognition/crosswalk_traffic_light_estimator/crosswalk_traffic_light_estimator.param.yaml]
  • launch/traffic_light_recognition/traffic_light_occlusion_predictor.launch.xml
    • Occlusion check for one traffic light camera: merges the car and pedestrian signals into the per-camera output, masking signals hidden from the camera.
      • perception_config_pkg [default: autoware_perception_config]
      • camera_namespace
      • input/cloud [default: /sensing/lidar/top/pointcloud_raw_ex]
      • input/vector_map [default: /map/vector_map]
      • occlusion_predictor_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_occlusion_predictor/traffic_light_occlusion_predictor.param.yaml]
  • launch/traffic_light_recognition/traffic_light_recognition_fine.launch.xml
    • Traffic light recognition, fine detection mode: per-camera fine detection with occlusion check, then shared fusion. Two cameras as the representative set; add or remove per-camera blocks to match the vehicle.
      • perception_config_pkg [default: autoware_perception_config]
      • input/vector_map [default: /map/vector_map]
      • input/route [default: /planning/mission_planning/route]
      • input/pointcloud [default: /sensing/lidar/top/pointcloud_raw_ex]
      • external/traffic_signals [default: /perception/traffic_light_recognition/external/traffic_signals]
      • output/traffic_signals [default: /perception/traffic_light_recognition/traffic_signals]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • fine_detector_model_path [default: $(var data_path)/traffic_light_fine_detector/tlr_car_ped_yolox_s_batch_6.onnx]
      • fine_detector_label_path [default: $(var data_path)/traffic_light_fine_detector/tlr_labels.txt]
      • lamp_recognizer_ml_param_path [default: $(var data_path)/traffic_light_classifier/lamp_recognizer_ml.param.yaml]
      • car_classifier_model_path [default: $(var data_path)/traffic_light_classifier/traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • car_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels.txt]
      • pedestrian_classifier_model_path [default: $(var data_path)/traffic_light_classifier/ped_traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • pedestrian_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels_ped.txt]
      • car_classifier_type [default: 1]
      • pedestrian_classifier_type [default: 1]
      • base_map_based_detector_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_map_based_detector/base_traffic_light_map_based_detector.param.yaml]
      • extra_map_based_detector_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_map_based_detector/extra_traffic_light_map_based_detector.param.yaml]
  • launch/traffic_light_recognition/traffic_light_recognition_merge.launch.xml
    • Traffic light recognition, merge mode: detection and classification run on other ECUs; this side runs the per-camera occlusion check and the shared fusion. Two cameras as the representative set; add or remove per-camera blocks to match the vehicle.
      • input/vector_map [default: /map/vector_map]
      • input/pointcloud [default: /sensing/lidar/top/pointcloud_raw_ex]
      • external/traffic_signals [default: /perception/traffic_light_recognition/external/traffic_signals]
      • output/traffic_signals [default: /perception/traffic_light_recognition/traffic_signals]
  • launch/traffic_light_recognition/traffic_light_recognition_rough.launch.xml
    • Traffic light recognition, rough detection mode: per-camera map based detection with occlusion check, then shared fusion. One camera as the representative set; add or remove per-camera blocks to match the vehicle.
      • perception_config_pkg [default: autoware_perception_config]
      • input/vector_map [default: /map/vector_map]
      • input/route [default: /planning/mission_planning/route]
      • input/pointcloud [default: /sensing/lidar/top/pointcloud_raw_ex]
      • external/traffic_signals [default: /perception/traffic_light_recognition/external/traffic_signals]
      • output/traffic_signals [default: /perception/traffic_light_recognition/traffic_signals]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • lamp_recognizer_ml_param_path [default: $(var data_path)/traffic_light_classifier/lamp_recognizer_ml.param.yaml]
      • car_classifier_model_path [default: $(var data_path)/traffic_light_classifier/traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • car_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels.txt]
      • pedestrian_classifier_model_path [default: $(var data_path)/traffic_light_classifier/ped_traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • pedestrian_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels_ped.txt]
      • car_classifier_type [default: 1]
      • pedestrian_classifier_type [default: 1]
      • base_map_based_detector_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_map_based_detector/base_traffic_light_map_based_detector.param.yaml]
  • launch/traffic_light_recognition/traffic_light_recognition_whole_image.launch.xml
    • Traffic light recognition, whole-image detection mode: per-camera yolox detection, then shared fusion. Two cameras as the representative set; add or remove per-camera blocks to match the vehicle.
      • perception_config_pkg [default: autoware_perception_config]
      • input/vector_map [default: /map/vector_map]
      • input/route [default: /planning/mission_planning/route]
      • external/traffic_signals [default: /perception/traffic_light_recognition/external/traffic_signals]
      • output/traffic_signals [default: /perception/traffic_light_recognition/traffic_signals]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • whole_image_detector_model_path [default: $(var data_path)/tensorrt_yolox/yolox_s_car_ped_tl_detector_960_960_batch_1.onnx]
      • whole_image_detector_label_path [default: $(var data_path)/tensorrt_yolox/car_ped_tl_detector_labels.txt]
      • lamp_recognizer_ml_param_path [default: $(var data_path)/traffic_light_classifier/lamp_recognizer_ml.param.yaml]
      • car_classifier_model_path [default: $(var data_path)/traffic_light_classifier/traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • car_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels.txt]
      • pedestrian_classifier_model_path [default: $(var data_path)/traffic_light_classifier/ped_traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • pedestrian_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels_ped.txt]
      • car_classifier_type [default: 1]
      • pedestrian_classifier_type [default: 1]
      • base_map_based_detector_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_map_based_detector/base_traffic_light_map_based_detector.param.yaml]
      • extra_map_based_detector_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_map_based_detector/extra_traffic_light_map_based_detector.param.yaml]

Messages

No message files found.

Services

No service files found

Plugins

No plugins found.

Recent questions tagged autoware_perception_launch at Robotics Stack Exchange

No version for distro jazzy showing github. Known supported distros are highlighted in the buttons above.
Package symbol

autoware_perception_launch package from autoware_launch repo

autoware_launch autoware_sample_designs autoware_control_config autoware_control_launch autoware_localization_config autoware_localization_launch autoware_map_config autoware_map_launch autoware_perception_config autoware_perception_launch autoware_planning_config autoware_planning_launch autoware_sensing_launch autoware_sensor_driver_launch awsim_labs_sensor_kit_description awsim_labs_sensor_kit_launch awsim_sensor_kit_description awsim_sensor_kit_launch carla_sensor_kit_description carla_sensor_kit_launch sample_sensor_kit_description sample_sensor_kit_launch single_lidar_sensor_kit_description single_lidar_sensor_kit_launch tier4_autoware_api_launch tier4_simulator_launch tier4_system_launch tier4_vehicle_launch awsim_labs_vehicle_description awsim_labs_vehicle_launch sample_vehicle_description sample_vehicle_launch

ROS Distro
github

API Docs
Browse Code

Package Summary

Version 0.53.0
License Apache License 2.0
Build type AMENT_CMAKE
Use RECOMMENDED

Repository Summary

Description
Checkout URI https://github.com/autowarefoundation/autoware_launch.git
VCS Type git
VCS Version main
Last Updated 2026-10-26
Dev Status UNKNOWN
Released UNRELEASED
Contributing Help Wanted (-)
Good First Issues (-)
Pull Requests to Review (-)

Package Description

The autoware_perception_launch package

Maintainers

  • Yukihiro Saito
  • Dai Nguyen
  • Taekjin Lee
  • Yoshi Ri
  • Masato Saeki

Authors

No additional authors.
autoware_universe_launch/autoware_perception_launch/README.md

autoware_perception_launch

Shared perception launch library. The system component launchers live in the product layer (autoware_launch/launch/components/component_perception.launch.xml for the lidar pipeline and component_traffic_light.launch.xml for traffic light recognition); this package hosts the perception feature launchers they combine:

launch/
├── common/                        # shared preprocess feature launchers (pointcloud downsample)
├── object_recognition/
│   ├── object_recognition_<mode>.launch.xml  # detection + tracking + prediction per sensor mode
│   ├── detection/
│   │   ├── detector/              # detectors
│   │   ├── filter/                # filters
│   │   └── merger/                # detection-side serial merger (used by the camera_lidar_radar_serial launcher)
│   └── prediction/
├── obstacle_segmentation/
├── occupancy_grid_map/
└── traffic_light_recognition/
design/module/                     # Autoware System Designer modules; same subtree as launch/
├── object_recognition/
├── obstacle_segmentation/
├── occupancy_grid_map/
└── traffic_light_recognition/

Launch files

  • object_recognition/object_recognition_<mode>.launch.xml — one feature launcher per sensor mode (camera, lidar, lidar_ptv3, lidar_radar, camera_lidar_radar, camera_lidar_radar_serial); hosts the detection stage inline together with the tracker and prediction, and pushes their module namespaces. Camera and radar input topics default to their absolute sensing paths (/sensing/camera/cameraN/..., /sensing/radar/detected_objects), so callers override only what differs. Most modes wire the multi-channel tracker to the detector outputs; camera_lidar_radar_serial instead merges the detectors in the detection stage (optionally regenerating detections through detection-by-tracker), tracks the single merged detected_objects channel, and joins radar tracked objects at the decorative tracker merger. camera runs camera-only detectors (BEV and camera VRU).
  • The clustering backend is file selection, not a flag: lidar runs rule-based euclidean clustering off the obstacle segmentation cloud, while lidar_ptv3 and lidar_radar cluster the PTv3 semantic-segmentation cloud and drop the pointcloud map filter and the object validator that only the rule-based pipeline uses. Both feed their cluster objects to the tracker on the lidar_clustering channel.
  • object_recognition/detection/{detector,filter}/ — single-purpose launchers wired together by the object-recognition launchers. merger/camera_lidar_merger.launch.xml chains the detection-side object mergers for the camera_lidar_radar_serial launcher.
  • obstacle_segmentation/ground_segmentation_<backend>.launch.xml — ground removal backends (rule_based, rule_based_cuda, ml); backend selection is file selection. ground_segmentation_{additional_lidars,ransac}.launch.xml fold extra lidars into the obstacle stream for kits that carry them.
  • occupancy_grid_map/, common/, traffic_light_recognition/ — feature launchers included directly by the system components.

Namespace and container addressing

Callers own the namespaces: the system component pushes perception and object_recognition; object_recognition_<mode>.launch.xml pushes detection/tracking/prediction. The object-recognition launchers address their internal topics through the fixed /perception/object_recognition namespace. Composable nodes load into the shared pointcloud container, whose name is exposed as pointcloud_container_name and aliased to the node/pointcloud_container argument of the detector launchers.

Config

Parameter files are resolved from a config/ tree addressed through the perception_config_pkg argument, which defaults to autoware_perception_config. A product can relocate the whole tree by setting perception_config_pkg once at its entry point — the launch configuration propagates through every include — but it must then ship the same config/ layout. Products that differ in only a few files instead override the corresponding *_param_path arguments, which every feature launcher declares with a default:

<include file="$(find-pkg-share autoware_perception_launch)/launch/object_recognition/object_recognition_lidar.launch.xml">
  <arg name="perception_config_pkg" value="my_perception_config"/>
  <arg name="FOO_NODE_param_path" value="..."/>
</include>

component_perception.launch.xml, component_traffic_light.launch.xml and planning_simulator/perception_simulation.launch.xml each declare perception_config_pkg, so a product switches the tree at the component it launches.

The launch/ tree is the interface downstream distributions adopt as-is; only package.xml, CMakeLists.txt, README.md, and changelogs stay package-local. Keep changes to this tree folder-merge compatible.

Package Dependencies

Please see <exec_depend> in package.xml.

CHANGELOG

Changelog for package autoware_perception_launch

0.53.0 (2026-10-13)

  • Merge remote-tracking branch 'origin/main' into tmp/bot/bump_version_base

  • feat(autoware_launch): run crop_box_filter standalone when agnocast is enabled (#1990)

    * feat(autoware_launch): run crop_box_filter standalone when agnocast is enabled autoware_universe#13384 moves CropBoxFilterComponent to agnocast_wrapper::Node on an agnocast-only executor, which the component containers do not set up. Start each of the eight crop_box_filter nodes as a standalone process when ENABLE_AGNOCAST=1 and keep loading them into their container otherwise. The node names, namespaces, remaps and parameters are the same in both branches.

    * refactor(autoware_launch): launch crop_box_filter through the autoware_node action ---------

  • feat(autoware_perception_launch): run scan_ground_filter standalone when agnocast is enabled (#1991)

    * feat(autoware_perception_launch): run scan_ground_filter standalone when agnocast is enabled autoware_universe#13385 moves ScanGroundFilterComponent to agnocast_wrapper::Node on an agnocast-only executor, which the component containers do not set up. Start each of the three scan_ground_filter nodes as a standalone process when ENABLE_AGNOCAST=1 and keep loading them into their container otherwise. The node names, namespaces, remaps and parameters are the same in both branches.

    * refactor(autoware_perception_launch): launch scan_ground_filter through the autoware_node action ---------

  • feat(autoware_launch): run voxel_grid_downsample_filter standalone when agnocast is enabled (#1992)

    • feat(autoware_launch): run voxel_grid_downsample_filter standalone when agnocast is enabled
    • style(pre-commit): autofix

    * refactor(autoware_launch): launch voxel_grid_downsample_filter through the autoware_node action ---------Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>

  • feat(autoware_sample_designs): compose the AD API, system and vehicle components from design modules (#1976)

    * fix(control, planning): connect the operation mode state and velocity limit inputs Wires collision_detector's operation_mode_state and remaining_distance_time_calculator's planning_velocity through the module designs; their node designs now declare the fixed topic names as remap targets.

    * fix(planning): connect the internal velocity limit path to the selector Forwards motion_velocity_planner's velocity_limit and clear_velocity_limit up to ScenarioPlanning and feeds them into external_velocity_limit_selector, whose internal inputs were unconnected.

    * fix(control, planning): connect the remaining module inputs to their sources Wires the operation mode transition manager's gate and selector inputs, the stop mode operator's vehicle status, the manual lane change handler's route availability and preferred primitive service, the parking costmap, and the route marker. Exposes the vehicle control mode service, the preferred lane service and the system hazard lights command as component interfaces.

    • fix(planning): follow the hazard lights selector input rename
    • fix(planning): add steering_status subscriber to various planning modules
    • feat(vehicle): add vehicle wrapper and launch file for vehicle interface
    • style(pre-commit): autofix
    • fix(planning): add planning factors publishers and connections for various modules
    • refactor(autoware_sample_designs): compose the AD API from design modules
    • Refactor ADAPI module structure and enhance planning factors
    • Updated AdapiRvizHelpers.module.yaml to include helper prefixes for instances and connections.
    • Modified AutowareSample.system.yaml to reorganize ADAPI components, introducing default_adapi for better structure.
    • Expanded planning factors in Planning.module.yaml, LaneDriving.module.yaml, ScenarioPlanning.module.yaml, and associated modules to include additional factors for improved behavior and motion planning.
    • Enhanced connections in behavior planning and motion planning modules to accommodate new planning factors.
    • Ensured consistency across scenario planning modules by aligning

File truncated at 100 lines see the full file

Package Dependencies

Deps Name
ament_cmake_auto
autoware_cmake
autoware_agnocast_wrapper
autoware_bevfusion
autoware_bytetrack
autoware_compare_map_segmentation
autoware_crosswalk_traffic_light_estimator
autoware_cuda_pointcloud_preprocessor
autoware_detected_object_feature_remover
autoware_detected_object_validation
autoware_detection_by_tracker
autoware_elevation_map_loader
autoware_euclidean_cluster
autoware_ground_segmentation
autoware_ground_segmentation_cuda
autoware_image_object_locator
autoware_image_projection_based_fusion
autoware_image_transport_decompressor
autoware_lidar_apollo_instance_segmentation
autoware_lidar_centerpoint
autoware_lidar_transfusion
autoware_map_based_prediction
autoware_multi_object_tracker
autoware_object_merger
autoware_object_sorter
autoware_occupancy_grid_map_outlier_filter
autoware_perception_config
autoware_pointcloud_preprocessor
autoware_probabilistic_occupancy_grid_map
autoware_ptv3
autoware_raindrop_cluster_filter
autoware_shape_estimation
autoware_simpl_prediction
autoware_simple_object_merger
autoware_tensorrt_bevdet
autoware_tensorrt_yolox
autoware_tracking_object_merger
autoware_traffic_light_arbiter
autoware_traffic_light_category_merger
autoware_traffic_light_classifier
autoware_traffic_light_fine_detector
autoware_traffic_light_map_based_detector
autoware_traffic_light_multi_camera_fusion
autoware_traffic_light_occlusion_predictor
autoware_traffic_light_selector
autoware_traffic_light_visualization
rclcpp_components
topic_tools
ament_lint_auto
autoware_lint_common

System Dependencies

No direct system dependencies.

Dependant Packages

Name Deps
autoware_launch
reaction_analyzer

Launch files

  • launch/common/pointcloud_downsample.launch.xml
      • pointcloud_container_name
      • cuda_pointcloud_preprocessing [default: false]
      • input/pointcloud
      • output/pointcloud
      • common_downsample_voxel_size_x [default: 0.05]
      • common_downsample_voxel_size_y [default: 0.05]
      • common_downsample_voxel_size_z [default: 0.05]
  • launch/object_recognition/detection/detector/camera_2d_detector.launch.xml
      • image_raw0 [default: /sensing/camera/camera0/image_raw]
      • image_raw1 [default: /sensing/camera/camera1/image_raw]
      • image_raw2 [default: /sensing/camera/camera2/image_raw]
      • image_raw3 [default: /sensing/camera/camera3/image_raw]
      • image_raw4 [default: /sensing/camera/camera4/image_raw]
      • image_raw5 [default: /sensing/camera/camera5/image_raw]
      • image_raw6 [default: /sensing/camera/camera6/image_raw]
      • image_raw7 [default: /sensing/camera/camera7/image_raw]
      • image_raw8 [default: /sensing/camera/camera8/image_raw]
      • image_raw9 [default: /sensing/camera/camera9/image_raw]
      • image_number [default: 1]
      • camera_index [default: 0]
      • use_bytetrack [default: true]
      • enable_visualizer [default: false]
      • camera_2d_detector/model_path
      • camera_2d_detector/label_path
      • camera_2d_detector/color_map_path
      • camera_2d_detector/roi_remap_path
      • camera_2d_detector/roi_to_semseg_remap_path
      • tensorrt_yolox_ns [default: ]
  • launch/object_recognition/detection/detector/camera_bev_detector.launch.xml
      • input/camera0/image
      • input/camera0/info
      • input/camera1/image
      • input/camera1/info
      • input/camera2/image
      • input/camera2/info
      • input/camera3/image
      • input/camera3/info
      • input/camera4/image
      • input/camera4/info
      • input/camera5/image
      • input/camera5/info
      • output/objects
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • bevdet_model_name [default: bevdet_one_lt_d]
      • bevdet_model_path [default: $(var data_path)/tensorrt_bevdet]
  • launch/object_recognition/detection/detector/camera_lidar_detector.launch.xml
      • ns
      • lidar_detection_model_type
      • lidar_detection_model_name
      • use_low_intensity_cluster_filter
      • use_image_segmentation_based_filter
      • number_of_cameras
      • input/camera0/image
      • input/camera0/info
      • input/camera0/rois
      • input/camera1/image
      • input/camera1/info
      • input/camera1/rois
      • input/camera2/image
      • input/camera2/info
      • input/camera2/rois
      • input/camera3/image
      • input/camera3/info
      • input/camera3/rois
      • input/camera4/image
      • input/camera4/info
      • input/camera4/rois
      • input/camera5/image
      • input/camera5/info
      • input/camera5/rois
      • input/camera6/image
      • input/camera6/info
      • input/camera6/rois
      • input/camera7/image
      • input/camera7/info
      • input/camera7/rois
      • input/camera8/image
      • input/camera8/info
      • input/camera8/rois
      • input/concatenation_info
      • segmentation_pointcloud_fusion_camera_ids
      • image_topic_name
      • sync_param_path
      • voxel_grid_based_euclidean_param_path
      • lidar_model_param_path
      • pointpainting_model_path [default: $(var data_path)/image_projection_based_fusion]
      • pointpainting_fusion_common_param_path
      • roi_cluster_fusion_param_path
      • node/pointcloud_container
      • input/pointcloud
      • input/pointcloud_map/pointcloud
      • input/obstacle_segmentation/pointcloud
      • output/ml_detector/objects
      • output/rule_detector/objects
      • camera_2d_detector/model_path
      • camera_2d_detector/label_path
      • camera_2d_detector/color_map_path
      • camera_2d_detector/roi_remap_path
      • camera_2d_detector/roi_to_semseg_remap_path
      • enable_2d_detection [default: false]
  • launch/object_recognition/detection/detector/camera_lidar_irregular_object_detector.launch.xml
      • pipeline_ns
      • node/pointcloud_container
      • output/objects [default: objects]
      • input/concatenation_info
      • input/pointcloud
      • irregular_object_crop_box_filter_node_param_path
      • irregular_object_scan_ground_filter_node_param_path
      • irregular_object_roi_pointcloud_fusion_node_param_path
      • irregular_object_fusion_camera_param_path
      • sync_param_path
  • launch/object_recognition/detection/detector/camera_vru_detector.launch.xml
      • input/camera0/info [default: /sensing/camera/camera0/camera_info]
      • input/camera0/rois [default: /perception/object_recognition/detection/rois0]
      • input/camera1/info [default: /sensing/camera/camera1/camera_info]
      • input/camera1/rois [default: /perception/object_recognition/detection/rois1]
      • input/camera2/info [default: /sensing/camera/camera2/camera_info]
      • input/camera2/rois [default: /perception/object_recognition/detection/rois2]
      • input/camera3/info [default: /sensing/camera/camera3/camera_info]
      • input/camera3/rois [default: /perception/object_recognition/detection/rois3]
      • input/camera4/info [default: /sensing/camera/camera4/camera_info]
      • input/camera4/rois [default: /perception/object_recognition/detection/rois4]
      • input/camera5/info [default: /sensing/camera/camera5/camera_info]
      • input/camera5/rois [default: /perception/object_recognition/detection/rois5]
      • input/camera6/info [default: /sensing/camera/camera6/camera_info]
      • input/camera6/rois [default: /perception/object_recognition/detection/rois6]
      • input/camera7/info [default: /sensing/camera/camera7/camera_info]
      • input/camera7/rois [default: /perception/object_recognition/detection/rois7]
      • output/objects [default: /perception/object_recognition/detection/camera_vru/objects]
      • bbox_object_locator_param_path
      • rois_ids [default: [0, 1]]
  • launch/object_recognition/detection/detector/irregular_object_detector.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • target_container
      • base_frame [default: base_link]
      • use_intra_process [default: true]
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • input/concatenation_info [default: /sensing/lidar/concatenated/pointcloud_info]
      • output/clusters
      • crop_box_filter_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/crop_box_filter.param.yaml]
      • scan_ground_filter_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/scan_ground_filter.param.yaml]
      • roi_pointcloud_fusion_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/roi_pointcloud_fusion.param.yaml]
      • fusion_camera_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/fusion_cameras.param.yaml]
      • sync_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/fusion_common.param.yaml]
  • launch/object_recognition/detection/detector/lidar_dnn_detector.launch.xml
      • lidar_detection_model_type
      • lidar_detection_model_name
      • lidar_model_param_path
      • bevfusion_model_path [default: $(var data_path)/bevfusion]
      • centerpoint_model_path [default: $(var data_path)/lidar_centerpoint]
      • centerpoint_common_param_path [default: $(find-pkg-share autoware_lidar_centerpoint)/config/centerpoint_common.param.yaml]
      • transfusion_model_path [default: $(var data_path)/lidar_transfusion]
      • use_short_range_detection [default: false]
      • lidar_short_range_detection_model_type
      • node/pointcloud_container
      • input/pointcloud
      • output/objects
      • output/short_range_objects
  • launch/object_recognition/detection/detector/lidar_rule_detector.launch.xml
      • ns
      • node/pointcloud_container
      • input/pointcloud_map/pointcloud
      • input/obstacle_segmentation/pointcloud
      • output/cluster_objects
      • output/objects
      • voxel_grid_based_euclidean_param_path
      • use_low_height_cropbox [default: false]
  • launch/object_recognition/detection/detector/lidar_semantic_segmentation_detector.launch.xml
      • euclidean_clustering/param_path
      • input/pointcloud
      • output/objects
  • launch/object_recognition/detection/detector/tracker_based_detector.launch.xml
      • detection_by_tracker_param_path
      • input/clusters
      • input/tracked_objects
      • output/objects
  • launch/object_recognition/detection/filter/object_filter.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • objects_filter_method [default: lanelet_filter]
      • object_recognition_detection_object_lanelet_filter_param_path [default: $(var perception_config_dir)/object_recognition/detection/object_filter/object_lanelet_filter.param.yaml]
      • object_recognition_detection_object_position_filter_param_path [default: $(var perception_config_dir)/object_recognition/detection/object_filter/object_position_filter.param.yaml]
      • input/objects
      • output/objects
  • launch/object_recognition/detection/filter/object_validator.launch.xml
      • objects_validation_method
      • object_recognition_detection_obstacle_pointcloud_based_validator_param_path
      • input/obstacle_pointcloud
      • input/objects
      • output/objects
  • launch/object_recognition/detection/filter/pointcloud_downsample_filter.launch.xml
      • target_container
      • use_intra_process [default: true]
      • input/pointcloud
      • output/pointcloud
      • downsample_voxel_size [default: 0.1]
  • launch/object_recognition/detection/filter/pointcloud_map_filter.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • target_container
      • use_intra_process [default: true]
      • input/pointcloud
      • output/pointcloud
      • output/downsampled/pointcloud [default: /perception/obstacle_segmentation/pointcloud_map_filtered/downsampled/pointcloud]
      • downsample_voxel_size [default: 0.1]
      • voxel_based_compare_map_filter_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/pointcloud_filter/voxel_based_compare_map_filter.param.yaml]
  • launch/object_recognition/detection/filter/radar_filter.launch.xml
      • object_sorter_param_path
      • radar_lanelet_filtering_range_param_path
      • input/radar
      • output/objects
  • launch/object_recognition/detection/merger/camera_lidar_merger.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • object_recognition_detection_object_merger_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/detection/object_merger/data_association_matrix.param.yaml]
      • object_recognition_detection_object_merger_distance_threshold_list_path [default: $(var perception_config_dir)/object_recognition/detection/object_merger/overlapped_judge.param.yaml]
      • ml_camera_lidar_object_association_merger_param_path [default: $(var perception_config_dir)/object_recognition/detection/object_merger/ml_camera_lidar_object_association_merger.param.yaml]
      • roi_detected_object_fusion_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/roi_detected_object_fusion.param.yaml]
      • sync_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/fusion_common.param.yaml]
      • lidar_detection_model_type
      • use_detection_by_tracker
      • use_irregular_object_detector
      • use_roi_detected_object_validator
      • euclidean_clustering/shape_policy
      • use_object_filter
      • objects_filter_method
      • number_of_cameras
      • input/camera0/image
      • input/camera0/info
      • input/camera0/rois
      • input/camera1/image
      • input/camera1/info
      • input/camera1/rois
      • input/camera2/image
      • input/camera2/info
      • input/camera2/rois
      • input/camera3/image
      • input/camera3/info
      • input/camera3/rois
      • input/camera4/image
      • input/camera4/info
      • input/camera4/rois
      • input/camera5/image
      • input/camera5/info
      • input/camera5/rois
      • input/camera6/image
      • input/camera6/info
      • input/camera6/rois
      • input/camera7/image
      • input/camera7/info
      • input/camera7/rois
      • input/camera8/image
      • input/camera8/info
      • input/camera8/rois
      • input/concatenation_info [default: /sensing/lidar/concatenated/pointcloud_info]
      • input/lidar_ml/objects
      • input/lidar_cluster/objects
      • input/irregular_object/objects [default: ]
      • input/detection_by_tracker/objects
      • output/objects [default: objects]
      • alpha_merger_priority_mode [default: 0]
  • launch/object_recognition/object_recognition_camera.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • use_camera_vru_detector [default: false]
      • camera_vru_detector_rois_ids [default: [0]]
      • use_camera_bev_detector [default: true]
      • use_vector_map [default: true]
      • tracker_publish_merged_objects [default: false]
      • input/camera0/image [default: /sensing/camera/camera0/image_rect_color]
      • input/camera0/info [default: /sensing/camera/camera0/camera_info]
      • input/camera0/rois [default: /perception/object_recognition/detection/rois0]
      • input/camera1/image [default: /sensing/camera/camera1/image_rect_color]
      • input/camera1/info [default: /sensing/camera/camera1/camera_info]
      • input/camera1/rois [default: /perception/object_recognition/detection/rois1]
      • input/camera2/image [default: /sensing/camera/camera2/image_rect_color]
      • input/camera2/info [default: /sensing/camera/camera2/camera_info]
      • input/camera2/rois [default: /perception/object_recognition/detection/rois2]
      • input/camera3/image [default: /sensing/camera/camera3/image_rect_color]
      • input/camera3/info [default: /sensing/camera/camera3/camera_info]
      • input/camera3/rois [default: /perception/object_recognition/detection/rois3]
      • input/camera4/image [default: /sensing/camera/camera4/image_rect_color]
      • input/camera4/info [default: /sensing/camera/camera4/camera_info]
      • input/camera4/rois [default: /perception/object_recognition/detection/rois4]
      • input/camera5/image [default: /sensing/camera/camera5/image_rect_color]
      • input/camera5/info [default: /sensing/camera/camera5/camera_info]
      • input/camera5/rois [default: /perception/object_recognition/detection/rois5]
      • input/camera6/image [default: /sensing/camera/camera6/image_rect_color]
      • input/camera6/info [default: /sensing/camera/camera6/camera_info]
      • input/camera6/rois [default: /perception/object_recognition/detection/rois6]
      • input/camera7/image [default: /sensing/camera/camera7/image_rect_color]
      • input/camera7/info [default: /sensing/camera/camera7/camera_info]
      • input/camera7/rois [default: /perception/object_recognition/detection/rois7]
      • output/tracker_merged_objects [default: /perception/object_recognition/detection/objects]
      • output/tracked_objects [default: /perception/object_recognition/tracking/objects]
      • output/predicted_objects [default: /perception/object_recognition/objects]
      • near_range_camera_vru_param_path [default: $(var perception_config_dir)/object_recognition/detection/camera_vru_detection/near_range_camera_vru_detector.param.yaml]
      • object_recognition_tracking_multi_object_tracker_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/data_association_matrix.param.yaml]
      • object_recognition_tracking_multi_object_tracker_input_channels_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/input_channels.param.yaml]
      • object_recognition_tracking_multi_object_tracker_node_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/multi_object_tracker_node.param.yaml]
  • launch/object_recognition/object_recognition_camera_lidar_radar.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • lidar_detection_model [default: centerpoint]
      • use_short_range_detection [default: false]
      • lidar_short_range_detection_model_type [default: centerpoint_short_range]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • bevfusion_model_path [default: $(var data_path)/bevfusion]
      • pointpainting_model_path [default: $(var data_path)/image_projection_based_fusion]
      • camera_2d_detector/model_path [default: $(var data_path)/tensorrt_yolox/yolox-sPlus-opt-pseudoV2-T4-960x960-T4-seg16cls.onnx]
      • camera_2d_detector/label_path [default: $(var data_path)/tensorrt_yolox/label.txt]
      • camera_2d_detector/color_map_path [default: $(var data_path)/tensorrt_yolox/semseg_color_map.csv]
      • camera_2d_detector/roi_remap_path [default: $(var perception_config_dir)/traffic_light_recognition/tensorrt_yolox/roi_label_remap.csv]
      • camera_2d_detector/roi_to_semseg_remap_path [default: $(var perception_config_dir)/traffic_light_recognition/tensorrt_yolox/roi_to_semseg_label_remap.csv]
      • use_semantic_segmentation_ptv3 [default: false]
      • use_object_filter [default: true]
      • use_pointcloud_map [default: true]
      • use_validator [default: false]
      • objects_validation_method [default: obstacle_pointcloud]
      • use_low_height_cropbox [default: false]
      • use_low_intensity_cluster_filter [default: true]
      • use_image_segmentation_based_filter [default: false]
      • use_irregular_object_detector [default: true]
      • use_camera_vru_detector [default: false]
      • camera_vru_detector_rois_ids [default: [0]]
      • use_camera_only_detection [default: false]
      • use_vector_map [default: true]
      • tracker_publish_merged_objects [default: false]
      • number_of_cameras [default: 6]
      • pointcloud_container_name [default: pointcloud_container]
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • input/semantic_segmentation/pointcloud [default: /perception/segmented/pointcloud]
      • input/obstacle_segmentation/pointcloud [default: /perception/obstacle_segmentation/pointcloud]
      • input/camera0/image [default: /sensing/camera/camera0/image_rect_color]
      • input/camera0/info [default: /sensing/camera/camera0/camera_info]
      • input/camera0/rois [default: /perception/object_recognition/detection/rois0]
      • input/camera1/image [default: /sensing/camera/camera1/image_rect_color]
      • input/camera1/info [default: /sensing/camera/camera1/camera_info]
      • input/camera1/rois [default: /perception/object_recognition/detection/rois1]
      • input/camera2/image [default: /sensing/camera/camera2/image_rect_color]
      • input/camera2/info [default: /sensing/camera/camera2/camera_info]
      • input/camera2/rois [default: /perception/object_recognition/detection/rois2]
      • input/camera3/image [default: /sensing/camera/camera3/image_rect_color]
      • input/camera3/info [default: /sensing/camera/camera3/camera_info]
      • input/camera3/rois [default: /perception/object_recognition/detection/rois3]
      • input/camera4/image [default: /sensing/camera/camera4/image_rect_color]
      • input/camera4/info [default: /sensing/camera/camera4/camera_info]
      • input/camera4/rois [default: /perception/object_recognition/detection/rois4]
      • input/camera5/image [default: /sensing/camera/camera5/image_rect_color]
      • input/camera5/info [default: /sensing/camera/camera5/camera_info]
      • input/camera5/rois [default: /perception/object_recognition/detection/rois5]
      • input/camera6/image [default: /sensing/camera/camera6/image_rect_color]
      • input/camera6/info [default: /sensing/camera/camera6/camera_info]
      • input/camera6/rois [default: /perception/object_recognition/detection/rois6]
      • input/camera7/image [default: /sensing/camera/camera7/image_rect_color]
      • input/camera7/info [default: /sensing/camera/camera7/camera_info]
      • input/camera7/rois [default: /perception/object_recognition/detection/rois7]
      • input/camera8/image [default: /sensing/camera/camera8/image_rect_color]
      • input/camera8/info [default: /sensing/camera/camera8/camera_info]
      • input/camera8/rois [default: /perception/object_recognition/detection/rois8]
      • input/concatenation_info [default: /sensing/lidar/concatenated/pointcloud_info]
      • input/radar [default: /sensing/radar/detected_objects]
      • input/camera_only/objects [default: /perception/object_recognition/detection/camera_only/objects]
      • input/camera_only/channel [default: camera_streampetr]
      • image_topic_name [default: image_rect_color]
      • segmentation_pointcloud_fusion_camera_ids [default: [0,2,4]]
      • output/tracker_merged_objects [default: /perception/object_recognition/detection/objects]
      • output/tracked_objects [default: /perception/object_recognition/tracking/objects]
      • output/predicted_objects [default: /perception/object_recognition/objects]
      • sync_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/fusion_common.param.yaml]
      • roi_cluster_fusion_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/roi_cluster_fusion.param.yaml]
      • pointpainting_fusion_common_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/pointpainting_common.param.yaml]
      • irregular_object_crop_box_filter_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/crop_box_filter.param.yaml]
      • irregular_object_scan_ground_filter_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/scan_ground_filter.param.yaml]
      • irregular_object_roi_pointcloud_fusion_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/roi_pointcloud_fusion.param.yaml]
      • irregular_object_fusion_camera_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/fusion_cameras.param.yaml]
      • lidar_model_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model]
      • centerpoint_common_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model/centerpoint_common.param.yaml]
      • voxel_grid_based_euclidean_param_path [default: $(var perception_config_dir)/object_recognition/detection/clustering/voxel_grid_based_euclidean_cluster.param.yaml]
      • radar_lanelet_filtering_range_param_path [default: $(var perception_config_dir)/object_recognition/detection/object_filter/radar_lanelet_filter.param.yaml]
      • object_recognition_detection_object_sorter_radar_param_path [default: $(var perception_config_dir)/object_recognition/detection/radar_filter/detected_object_sorter.param.yaml]
      • near_range_camera_vru_param_path [default: $(var perception_config_dir)/object_recognition/detection/camera_vru_detection/near_range_camera_vru_detector.param.yaml]
      • object_recognition_detection_obstacle_pointcloud_based_validator_param_path [default: $(var perception_config_dir)/object_recognition/detection/detected_object_validation/obstacle_pointcloud_based_validator.param.yaml]
      • object_recognition_tracking_multi_object_tracker_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/data_association_matrix.param.yaml]
      • object_recognition_tracking_multi_object_tracker_input_channels_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/input_channels.param.yaml]
      • object_recognition_tracking_multi_object_tracker_node_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/multi_object_tracker_node.param.yaml]
      • euclidean_clustering/param_path [default: $(var perception_config_dir)/object_recognition/detection/euclidean_cluster/label_based_euclidean_cluster.param.yaml]
  • launch/object_recognition/object_recognition_camera_lidar_radar_serial.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • lidar_detection_model [default: centerpoint]
      • use_short_range_detection [default: false]
      • lidar_short_range_detection_model_type [default: centerpoint_short_range]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • bevfusion_model_path [default: $(var data_path)/bevfusion]
      • pointpainting_model_path [default: $(var data_path)/image_projection_based_fusion]
      • camera_2d_detector/model_path [default: $(var data_path)/tensorrt_yolox/yolox-sPlus-opt-pseudoV2-T4-960x960-T4-seg16cls.onnx]
      • camera_2d_detector/label_path [default: $(var data_path)/tensorrt_yolox/label.txt]
      • camera_2d_detector/color_map_path [default: $(var data_path)/tensorrt_yolox/semseg_color_map.csv]
      • camera_2d_detector/roi_remap_path [default: $(var perception_config_dir)/traffic_light_recognition/tensorrt_yolox/roi_label_remap.csv]
      • camera_2d_detector/roi_to_semseg_remap_path [default: $(var perception_config_dir)/traffic_light_recognition/tensorrt_yolox/roi_to_semseg_label_remap.csv]
      • use_object_filter [default: true]
      • objects_filter_method [default: lanelet_filter]
      • use_pointcloud_map [default: true]
      • use_validator [default: false]
      • objects_validation_method [default: obstacle_pointcloud]
      • use_roi_detected_object_validator [default: true]
      • use_detection_by_tracker [default: true]
      • use_low_height_cropbox [default: false]
      • use_low_intensity_cluster_filter [default: true]
      • use_image_segmentation_based_filter [default: false]
      • use_irregular_object_detector [default: true]
      • ml_camera_lidar_merger_priority_mode [default: 0]
      • use_vector_map [default: true]
      • number_of_cameras [default: 6]
      • pointcloud_container_name [default: pointcloud_container]
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • input/obstacle_segmentation/pointcloud [default: /perception/obstacle_segmentation/pointcloud]
      • input/camera0/image [default: /sensing/camera/camera0/image_rect_color]
      • input/camera0/info [default: /sensing/camera/camera0/camera_info]
      • input/camera0/rois [default: /perception/object_recognition/detection/rois0]
      • input/camera1/image [default: /sensing/camera/camera1/image_rect_color]
      • input/camera1/info [default: /sensing/camera/camera1/camera_info]
      • input/camera1/rois [default: /perception/object_recognition/detection/rois1]
      • input/camera2/image [default: /sensing/camera/camera2/image_rect_color]
      • input/camera2/info [default: /sensing/camera/camera2/camera_info]
      • input/camera2/rois [default: /perception/object_recognition/detection/rois2]
      • input/camera3/image [default: /sensing/camera/camera3/image_rect_color]
      • input/camera3/info [default: /sensing/camera/camera3/camera_info]
      • input/camera3/rois [default: /perception/object_recognition/detection/rois3]
      • input/camera4/image [default: /sensing/camera/camera4/image_rect_color]
      • input/camera4/info [default: /sensing/camera/camera4/camera_info]
      • input/camera4/rois [default: /perception/object_recognition/detection/rois4]
      • input/camera5/image [default: /sensing/camera/camera5/image_rect_color]
      • input/camera5/info [default: /sensing/camera/camera5/camera_info]
      • input/camera5/rois [default: /perception/object_recognition/detection/rois5]
      • input/camera6/image [default: /sensing/camera/camera6/image_rect_color]
      • input/camera6/info [default: /sensing/camera/camera6/camera_info]
      • input/camera6/rois [default: /perception/object_recognition/detection/rois6]
      • input/camera7/image [default: /sensing/camera/camera7/image_rect_color]
      • input/camera7/info [default: /sensing/camera/camera7/camera_info]
      • input/camera7/rois [default: /perception/object_recognition/detection/rois7]
      • input/camera8/image [default: /sensing/camera/camera8/image_rect_color]
      • input/camera8/info [default: /sensing/camera/camera8/camera_info]
      • input/camera8/rois [default: /perception/object_recognition/detection/rois8]
      • input/concatenation_info [default: /sensing/lidar/concatenated/pointcloud_info]
      • input/radar/tracked_objects [default: /sensing/radar/tracked_objects]
      • image_topic_name [default: image_rect_color]
      • segmentation_pointcloud_fusion_camera_ids [default: [0,2,4]]
      • output/tracked_objects [default: /perception/object_recognition/tracking/objects]
      • output/predicted_objects [default: /perception/object_recognition/objects]
      • sync_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/fusion_common.param.yaml]
      • roi_cluster_fusion_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/roi_cluster_fusion.param.yaml]
      • pointpainting_fusion_common_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/pointpainting_common.param.yaml]
      • irregular_object_crop_box_filter_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/crop_box_filter.param.yaml]
      • irregular_object_scan_ground_filter_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/scan_ground_filter.param.yaml]
      • irregular_object_roi_pointcloud_fusion_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/roi_pointcloud_fusion.param.yaml]
      • irregular_object_fusion_camera_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/fusion_cameras.param.yaml]
      • lidar_model_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model]
      • centerpoint_common_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model/centerpoint_common.param.yaml]
      • voxel_grid_based_euclidean_param_path [default: $(var perception_config_dir)/object_recognition/detection/clustering/voxel_grid_based_euclidean_cluster.param.yaml]
      • detection_by_tracker_param_path [default: $(var perception_config_dir)/object_recognition/detection/detection_by_tracker/detection_by_tracker.param.yaml]
      • object_recognition_detection_obstacle_pointcloud_based_validator_param_path [default: $(var perception_config_dir)/object_recognition/detection/detected_object_validation/obstacle_pointcloud_based_validator.param.yaml]
      • object_recognition_detection_object_merger_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/detection/object_merger/data_association_matrix.param.yaml]
      • object_recognition_detection_object_merger_distance_threshold_list_path [default: $(var perception_config_dir)/object_recognition/detection/object_merger/overlapped_judge.param.yaml]
      • ml_camera_lidar_object_association_merger_param_path [default: $(var perception_config_dir)/object_recognition/detection/object_merger/ml_camera_lidar_object_association_merger.param.yaml]
      • roi_detected_object_fusion_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/roi_detected_object_fusion.param.yaml]
      • object_recognition_tracking_multi_object_tracker_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/data_association_matrix.param.yaml]
      • object_recognition_tracking_multi_object_tracker_input_channels_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/input_channels.param.yaml]
      • object_recognition_tracking_multi_object_tracker_node_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/multi_object_tracker_node.param.yaml]
      • object_recognition_tracking_radar_tracked_object_sorter_param_path [default: $(var perception_config_dir)/object_recognition/tracking/radar_object_tracker/tracked_object_sorter.param.yaml]
      • object_recognition_tracking_radar_tracked_object_lanelet_filter_param_path [default: $(var perception_config_dir)/object_recognition/tracking/radar_object_tracker/tracked_object_lanelet_filter.param.yaml]
      • object_recognition_tracking_object_merger_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/tracking/tracking_object_merger/data_association_matrix.param.yaml]
      • object_recognition_tracking_object_merger_node_param_path [default: $(var perception_config_dir)/object_recognition/tracking/tracking_object_merger/decorative_tracker_merger.param.yaml]
      • euclidean_clustering/shape_policy [default: 0]
  • launch/object_recognition/object_recognition_lidar.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • lidar_detection_model [default: centerpoint]
      • use_short_range_detection [default: false]
      • lidar_short_range_detection_model_type [default: centerpoint_short_range]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • bevfusion_model_path [default: $(var data_path)/bevfusion]
      • use_pointcloud_map [default: true]
      • use_validator [default: false]
      • objects_validation_method [default: obstacle_pointcloud]
      • use_object_filter [default: true]
      • objects_filter_method [default: lanelet_filter]
      • use_low_height_cropbox [default: false]
      • use_vector_map [default: true]
      • tracker_publish_merged_objects [default: false]
      • pointcloud_container_name [default: pointcloud_container]
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • input/obstacle_segmentation/pointcloud [default: /perception/obstacle_segmentation/pointcloud]
      • output/tracker_merged_objects [default: /perception/object_recognition/detection/objects]
      • output/tracked_objects [default: /perception/object_recognition/tracking/objects]
      • output/predicted_objects [default: /perception/object_recognition/objects]
      • lidar_model_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model]
      • centerpoint_common_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model/centerpoint_common.param.yaml]
      • voxel_grid_based_euclidean_param_path [default: $(var perception_config_dir)/object_recognition/detection/clustering/voxel_grid_based_euclidean_cluster.param.yaml]
      • object_recognition_detection_obstacle_pointcloud_based_validator_param_path [default: $(var perception_config_dir)/object_recognition/detection/detected_object_validation/obstacle_pointcloud_based_validator.param.yaml]
      • object_recognition_tracking_multi_object_tracker_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/data_association_matrix.param.yaml]
      • object_recognition_tracking_multi_object_tracker_input_channels_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/input_channels.param.yaml]
      • object_recognition_tracking_multi_object_tracker_node_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/multi_object_tracker_node.param.yaml]
  • launch/object_recognition/object_recognition_lidar_ptv3.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • lidar_detection_model [default: centerpoint]
      • use_short_range_detection [default: false]
      • lidar_short_range_detection_model_type [default: centerpoint_short_range]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • bevfusion_model_path [default: $(var data_path)/bevfusion]
      • use_vector_map [default: true]
      • tracker_publish_merged_objects [default: false]
      • pointcloud_container_name [default: pointcloud_container]
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • input/semantic_segmentation/pointcloud [default: /perception/segmented/pointcloud]
      • output/tracker_merged_objects [default: /perception/object_recognition/detection/objects]
      • output/tracked_objects [default: /perception/object_recognition/tracking/objects]
      • output/predicted_objects [default: /perception/object_recognition/objects]
      • lidar_model_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model]
      • centerpoint_common_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model/centerpoint_common.param.yaml]
      • object_recognition_tracking_multi_object_tracker_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/data_association_matrix.param.yaml]
      • object_recognition_tracking_multi_object_tracker_input_channels_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/input_channels.param.yaml]
      • object_recognition_tracking_multi_object_tracker_node_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/multi_object_tracker_node.param.yaml]
      • euclidean_clustering/param_path [default: $(var perception_config_dir)/object_recognition/detection/euclidean_cluster/label_based_euclidean_cluster.param.yaml]
  • launch/object_recognition/object_recognition_lidar_radar.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • lidar_detection_model [default: centerpoint]
      • use_short_range_detection [default: false]
      • lidar_short_range_detection_model_type [default: centerpoint_short_range]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • bevfusion_model_path [default: $(var data_path)/bevfusion]
      • use_low_height_cropbox [default: false]
      • use_vector_map [default: true]
      • tracker_publish_merged_objects [default: false]
      • pointcloud_container_name [default: pointcloud_container]
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • input/semantic_segmentation/pointcloud [default: /perception/segmented/pointcloud]
      • input/radar [default: /sensing/radar/detected_objects]
      • output/tracker_merged_objects [default: /perception/object_recognition/detection/objects]
      • output/tracked_objects [default: /perception/object_recognition/tracking/objects]
      • output/predicted_objects [default: /perception/object_recognition/objects]
      • lidar_model_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model]
      • centerpoint_common_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model/centerpoint_common.param.yaml]
      • voxel_grid_based_euclidean_param_path [default: $(var perception_config_dir)/object_recognition/detection/clustering/voxel_grid_based_euclidean_cluster.param.yaml]
      • radar_lanelet_filtering_range_param_path [default: $(var perception_config_dir)/object_recognition/detection/object_filter/radar_lanelet_filter.param.yaml]
      • object_recognition_detection_object_sorter_radar_param_path [default: $(var perception_config_dir)/object_recognition/detection/radar_filter/detected_object_sorter.param.yaml]
      • object_recognition_tracking_multi_object_tracker_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/data_association_matrix.param.yaml]
      • object_recognition_tracking_multi_object_tracker_input_channels_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/input_channels.param.yaml]
      • object_recognition_tracking_multi_object_tracker_node_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/multi_object_tracker_node.param.yaml]
      • euclidean_clustering/param_path [default: $(var perception_config_dir)/object_recognition/detection/euclidean_cluster/label_based_euclidean_cluster.param.yaml]
  • launch/object_recognition/prediction/prediction.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • use_vector_map [default: false]
      • prediction_model_type [default: map_based]
      • input/objects [default: /perception/object_recognition/tracking/objects]
      • output/objects [default: /perception/object_recognition/objects]
      • object_recognition_prediction_map_based_prediction_param_path [default: $(var perception_config_dir)/object_recognition/prediction/map_based_prediction.param.yaml]
  • launch/obstacle_segmentation/ground_segmentation_additional_lidars.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • target_container
      • lidar_name
      • base_frame [default: base_link]
      • use_intra_process [default: true]
      • input/main/pointcloud
      • input/lidar/pointcloud [default: /sensing/lidar/$(var lidar_name)/pointcloud]
      • output/pointcloud
      • output/pointcloud_info [default: $(var output/pointcloud)_info]
      • crop_box_margin_max_z [default: 2.5]
      • crop_box_margin_min_z [default: -2.5]
      • crop_box_filter_node_param_path [default: $(var perception_config_dir)/obstacle_segmentation/ground_segmentation/crop_box_filter.param.yaml]
      • scan_ground_filter_node_param_path [default: $(var perception_config_dir)/obstacle_segmentation/ground_segmentation/scan_ground_filter.param.yaml]
  • launch/obstacle_segmentation/ground_segmentation_ml.launch.xml
      • pointcloud_container_name [default: pointcloud_container]
      • use_pointcloud_container [default: true]
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • output/pointcloud [default: /perception/obstacle_segmentation/pointcloud]
      • output/pointcloud/segmentation [default: /perception/segmented/pointcloud]
      • output/pointcloud/visualization [default: /perception/segmented/pointcloud/visualization]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • model_variant [default: ptv3]
      • model_path [default: $(var data_path)/ptv3]
  • launch/obstacle_segmentation/ground_segmentation_ransac.launch.xml
      • target_container
      • base_frame [default: base_link]
      • use_intra_process [default: true]
      • input_topics
      • input/main/pointcloud
      • output/pointcloud
      • output/pointcloud_info [default: $(var output/pointcloud)_info]
      • detection_area_crop_box_filter_node_param_path
      • ransac_ground_filter_node_param_path [default: $(find-pkg-share autoware_ground_segmentation)/config/ransac_ground_filter.param.yaml]
  • launch/obstacle_segmentation/ground_segmentation_rule_based.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • target_container
      • base_frame [default: base_link]
      • use_intra_process [default: true]
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • output/pointcloud [default: /perception/obstacle_segmentation/pointcloud]
      • crop_box_margin_max_z [default: 2.5]
      • crop_box_margin_min_z [default: -2.5]
      • crop_box_filter_node_param_path [default: $(var perception_config_dir)/obstacle_segmentation/ground_segmentation/crop_box_filter.param.yaml]
      • scan_ground_filter_node_param_path [default: $(var perception_config_dir)/obstacle_segmentation/ground_segmentation/scan_ground_filter.param.yaml]
  • launch/obstacle_segmentation/ground_segmentation_rule_based_cuda.launch.xml
      • target_container
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • output/pointcloud [default: /perception/obstacle_segmentation/pointcloud]
      • output/ground_pointcloud [default: /perception/obstacle_segmentation/ground_pointcloud]
      • cuda_scan_ground_segmentation_filter_node_param_path [default: $(find-pkg-share autoware_ground_segmentation_cuda)/config/cuda_scan_ground_segmentation_filter.param.yaml]
  • launch/occupancy_grid_map/probabilistic_occupancy_grid_map.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • input/obstacle_pointcloud [default: concatenated/pointcloud]
      • input/raw_pointcloud [default: no_ground/oneshot/pointcloud]
      • output [default: /perception/occupancy_grid_map/map]
      • use_intra_process [default: false]
      • use_multithread [default: false]
      • pointcloud_container_name [default: pointcloud_container]
      • occupancy_grid_map_method
      • occupancy_grid_map_param_path [default: $(var perception_config_dir)/occupancy_grid_map/$(var occupancy_grid_map_method).param.yaml]
      • occupancy_grid_map_updater
      • occupancy_grid_map_updater_param_path [default: $(var perception_config_dir)/occupancy_grid_map/$(var occupancy_grid_map_updater)_updater.param.yaml]
      • input_obstacle_pointcloud [default: false]
      • input_obstacle_and_raw_pointcloud [default: true]
      • use_pointcloud_container [default: true]
  • launch/traffic_light_recognition/traffic_light_detector_fine.launch.xml
    • Fine-detection chain for one traffic light camera: map based detector proposes rough ROIs, the fine detector refines them, classifiers label them.
      • perception_config_pkg [default: autoware_perception_config]
      • camera_namespace
      • enable_image_decompressor [default: true]
      • input/vector_map [default: /map/vector_map]
      • input/route [default: /planning/mission_planning/route]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • map_based_detector_param_path [default: $(find-pkg-share autoware_traffic_light_map_based_detector)/config/traffic_light_map_based_detector.param.yaml]
      • fine_detector_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_fine_detector/traffic_light_fine_detector.param.yaml]
      • car_classifier_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_classifier/car_traffic_light_classifier.param.yaml]
      • pedestrian_classifier_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_classifier/pedestrian_traffic_light_classifier.param.yaml]
      • lamp_recognizer_ml_param_path [default: $(var data_path)/traffic_light_classifier/lamp_recognizer_ml.param.yaml]
      • roi_visualizer_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_visualization/traffic_light_roi_visualizer.param.yaml]
      • fine_detector_model_path [default: $(var data_path)/traffic_light_fine_detector/tlr_car_ped_yolox_s_batch_6.onnx]
      • fine_detector_label_path [default: $(var data_path)/traffic_light_fine_detector/tlr_labels.txt]
      • car_classifier_model_path [default: $(var data_path)/traffic_light_classifier/traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • car_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels.txt]
      • pedestrian_classifier_model_path [default: $(var data_path)/traffic_light_classifier/ped_traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • pedestrian_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels_ped.txt]
      • car_classifier_type [default: 1]
      • pedestrian_classifier_type [default: 1]
  • launch/traffic_light_recognition/traffic_light_detector_rough.launch.xml
    • Rough-detection chain for one traffic light camera: map based detector ROIs feed the classifiers directly, without a high accuracy detector.
      • perception_config_pkg [default: autoware_perception_config]
      • camera_namespace
      • enable_image_decompressor [default: true]
      • input/vector_map [default: /map/vector_map]
      • input/route [default: /planning/mission_planning/route]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • map_based_detector_param_path [default: $(find-pkg-share autoware_traffic_light_map_based_detector)/config/traffic_light_map_based_detector.param.yaml]
      • car_classifier_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_classifier/car_traffic_light_classifier.param.yaml]
      • pedestrian_classifier_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_classifier/pedestrian_traffic_light_classifier.param.yaml]
      • lamp_recognizer_ml_param_path [default: $(var data_path)/traffic_light_classifier/lamp_recognizer_ml.param.yaml]
      • roi_visualizer_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_visualization/traffic_light_roi_visualizer.param.yaml]
      • car_classifier_model_path [default: $(var data_path)/traffic_light_classifier/traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • car_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels.txt]
      • pedestrian_classifier_model_path [default: $(var data_path)/traffic_light_classifier/ped_traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • pedestrian_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels_ped.txt]
      • car_classifier_type [default: 1]
      • pedestrian_classifier_type [default: 1]
  • launch/traffic_light_recognition/traffic_light_detector_whole_image.launch.xml
    • Whole-image detection chain for one traffic light camera: yolox detects over the full image, the selector matches detections to map based ROIs, classifiers label them.
      • perception_config_pkg [default: autoware_perception_config]
      • camera_namespace
      • enable_image_decompressor [default: true]
      • input/vector_map [default: /map/vector_map]
      • input/route [default: /planning/mission_planning/route]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • map_based_detector_param_path [default: $(find-pkg-share autoware_traffic_light_map_based_detector)/config/traffic_light_map_based_detector.param.yaml]
      • whole_image_detector_param_path [default: $(var perception_config_dir)/traffic_light_recognition/tensorrt_yolox/yolox_traffic_light_detector.param.yaml]
      • car_classifier_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_classifier/car_traffic_light_classifier.param.yaml]
      • pedestrian_classifier_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_classifier/pedestrian_traffic_light_classifier.param.yaml]
      • lamp_recognizer_ml_param_path [default: $(var data_path)/traffic_light_classifier/lamp_recognizer_ml.param.yaml]
      • roi_visualizer_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_visualization/traffic_light_roi_visualizer.param.yaml]
      • whole_image_detector_model_path [default: $(var data_path)/tensorrt_yolox/yolox_s_car_ped_tl_detector_960_960_batch_1.onnx]
      • whole_image_detector_label_path [default: $(var data_path)/tensorrt_yolox/car_ped_tl_detector_labels.txt]
      • car_classifier_model_path [default: $(var data_path)/traffic_light_classifier/traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • car_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels.txt]
      • pedestrian_classifier_model_path [default: $(var data_path)/traffic_light_classifier/ped_traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • pedestrian_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels_ped.txt]
      • car_classifier_type [default: 1]
      • pedestrian_classifier_type [default: 1]
  • launch/traffic_light_recognition/traffic_light_fusion.launch.xml
    • Post-process shared by all traffic light cameras: multi camera fusion, V2X arbitration, pedestrian signal estimation and map visualization.
      • perception_config_pkg [default: autoware_perception_config]
      • camera_namespaces [default: [camera6, camera7]]
      • input/vector_map [default: /map/vector_map]
      • external/traffic_signals [default: /perception/traffic_light_recognition/external/traffic_signals]
      • internal/traffic_signals [default: /perception/traffic_light_recognition/internal/traffic_signals]
      • judged/traffic_signals [default: /perception/traffic_light_recognition/judged/traffic_signals]
      • output/traffic_signals [default: /perception/traffic_light_recognition/traffic_signals]
      • multi_camera_fusion_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_multi_camera_fusion/traffic_light_multi_camera_fusion.param.yaml]
      • arbiter_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_arbiter/traffic_light_arbiter.param.yaml]
      • crosswalk_estimator_param_path [default: $(var perception_config_dir)/traffic_light_recognition/crosswalk_traffic_light_estimator/crosswalk_traffic_light_estimator.param.yaml]
  • launch/traffic_light_recognition/traffic_light_occlusion_predictor.launch.xml
    • Occlusion check for one traffic light camera: merges the car and pedestrian signals into the per-camera output, masking signals hidden from the camera.
      • perception_config_pkg [default: autoware_perception_config]
      • camera_namespace
      • input/cloud [default: /sensing/lidar/top/pointcloud_raw_ex]
      • input/vector_map [default: /map/vector_map]
      • occlusion_predictor_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_occlusion_predictor/traffic_light_occlusion_predictor.param.yaml]
  • launch/traffic_light_recognition/traffic_light_recognition_fine.launch.xml
    • Traffic light recognition, fine detection mode: per-camera fine detection with occlusion check, then shared fusion. Two cameras as the representative set; add or remove per-camera blocks to match the vehicle.
      • perception_config_pkg [default: autoware_perception_config]
      • input/vector_map [default: /map/vector_map]
      • input/route [default: /planning/mission_planning/route]
      • input/pointcloud [default: /sensing/lidar/top/pointcloud_raw_ex]
      • external/traffic_signals [default: /perception/traffic_light_recognition/external/traffic_signals]
      • output/traffic_signals [default: /perception/traffic_light_recognition/traffic_signals]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • fine_detector_model_path [default: $(var data_path)/traffic_light_fine_detector/tlr_car_ped_yolox_s_batch_6.onnx]
      • fine_detector_label_path [default: $(var data_path)/traffic_light_fine_detector/tlr_labels.txt]
      • lamp_recognizer_ml_param_path [default: $(var data_path)/traffic_light_classifier/lamp_recognizer_ml.param.yaml]
      • car_classifier_model_path [default: $(var data_path)/traffic_light_classifier/traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • car_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels.txt]
      • pedestrian_classifier_model_path [default: $(var data_path)/traffic_light_classifier/ped_traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • pedestrian_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels_ped.txt]
      • car_classifier_type [default: 1]
      • pedestrian_classifier_type [default: 1]
      • base_map_based_detector_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_map_based_detector/base_traffic_light_map_based_detector.param.yaml]
      • extra_map_based_detector_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_map_based_detector/extra_traffic_light_map_based_detector.param.yaml]
  • launch/traffic_light_recognition/traffic_light_recognition_merge.launch.xml
    • Traffic light recognition, merge mode: detection and classification run on other ECUs; this side runs the per-camera occlusion check and the shared fusion. Two cameras as the representative set; add or remove per-camera blocks to match the vehicle.
      • input/vector_map [default: /map/vector_map]
      • input/pointcloud [default: /sensing/lidar/top/pointcloud_raw_ex]
      • external/traffic_signals [default: /perception/traffic_light_recognition/external/traffic_signals]
      • output/traffic_signals [default: /perception/traffic_light_recognition/traffic_signals]
  • launch/traffic_light_recognition/traffic_light_recognition_rough.launch.xml
    • Traffic light recognition, rough detection mode: per-camera map based detection with occlusion check, then shared fusion. One camera as the representative set; add or remove per-camera blocks to match the vehicle.
      • perception_config_pkg [default: autoware_perception_config]
      • input/vector_map [default: /map/vector_map]
      • input/route [default: /planning/mission_planning/route]
      • input/pointcloud [default: /sensing/lidar/top/pointcloud_raw_ex]
      • external/traffic_signals [default: /perception/traffic_light_recognition/external/traffic_signals]
      • output/traffic_signals [default: /perception/traffic_light_recognition/traffic_signals]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • lamp_recognizer_ml_param_path [default: $(var data_path)/traffic_light_classifier/lamp_recognizer_ml.param.yaml]
      • car_classifier_model_path [default: $(var data_path)/traffic_light_classifier/traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • car_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels.txt]
      • pedestrian_classifier_model_path [default: $(var data_path)/traffic_light_classifier/ped_traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • pedestrian_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels_ped.txt]
      • car_classifier_type [default: 1]
      • pedestrian_classifier_type [default: 1]
      • base_map_based_detector_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_map_based_detector/base_traffic_light_map_based_detector.param.yaml]
  • launch/traffic_light_recognition/traffic_light_recognition_whole_image.launch.xml
    • Traffic light recognition, whole-image detection mode: per-camera yolox detection, then shared fusion. Two cameras as the representative set; add or remove per-camera blocks to match the vehicle.
      • perception_config_pkg [default: autoware_perception_config]
      • input/vector_map [default: /map/vector_map]
      • input/route [default: /planning/mission_planning/route]
      • external/traffic_signals [default: /perception/traffic_light_recognition/external/traffic_signals]
      • output/traffic_signals [default: /perception/traffic_light_recognition/traffic_signals]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • whole_image_detector_model_path [default: $(var data_path)/tensorrt_yolox/yolox_s_car_ped_tl_detector_960_960_batch_1.onnx]
      • whole_image_detector_label_path [default: $(var data_path)/tensorrt_yolox/car_ped_tl_detector_labels.txt]
      • lamp_recognizer_ml_param_path [default: $(var data_path)/traffic_light_classifier/lamp_recognizer_ml.param.yaml]
      • car_classifier_model_path [default: $(var data_path)/traffic_light_classifier/traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • car_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels.txt]
      • pedestrian_classifier_model_path [default: $(var data_path)/traffic_light_classifier/ped_traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • pedestrian_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels_ped.txt]
      • car_classifier_type [default: 1]
      • pedestrian_classifier_type [default: 1]
      • base_map_based_detector_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_map_based_detector/base_traffic_light_map_based_detector.param.yaml]
      • extra_map_based_detector_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_map_based_detector/extra_traffic_light_map_based_detector.param.yaml]

Messages

No message files found.

Services

No service files found

Plugins

No plugins found.

Recent questions tagged autoware_perception_launch at Robotics Stack Exchange

No version for distro kilted showing github. Known supported distros are highlighted in the buttons above.
Package symbol

autoware_perception_launch package from autoware_launch repo

autoware_launch autoware_sample_designs autoware_control_config autoware_control_launch autoware_localization_config autoware_localization_launch autoware_map_config autoware_map_launch autoware_perception_config autoware_perception_launch autoware_planning_config autoware_planning_launch autoware_sensing_launch autoware_sensor_driver_launch awsim_labs_sensor_kit_description awsim_labs_sensor_kit_launch awsim_sensor_kit_description awsim_sensor_kit_launch carla_sensor_kit_description carla_sensor_kit_launch sample_sensor_kit_description sample_sensor_kit_launch single_lidar_sensor_kit_description single_lidar_sensor_kit_launch tier4_autoware_api_launch tier4_simulator_launch tier4_system_launch tier4_vehicle_launch awsim_labs_vehicle_description awsim_labs_vehicle_launch sample_vehicle_description sample_vehicle_launch

ROS Distro
github

API Docs
Browse Code

Package Summary

Version 0.53.0
License Apache License 2.0
Build type AMENT_CMAKE
Use RECOMMENDED

Repository Summary

Description
Checkout URI https://github.com/autowarefoundation/autoware_launch.git
VCS Type git
VCS Version main
Last Updated 2026-10-26
Dev Status UNKNOWN
Released UNRELEASED
Contributing Help Wanted (-)
Good First Issues (-)
Pull Requests to Review (-)

Package Description

The autoware_perception_launch package

Maintainers

  • Yukihiro Saito
  • Dai Nguyen
  • Taekjin Lee
  • Yoshi Ri
  • Masato Saeki

Authors

No additional authors.
autoware_universe_launch/autoware_perception_launch/README.md

autoware_perception_launch

Shared perception launch library. The system component launchers live in the product layer (autoware_launch/launch/components/component_perception.launch.xml for the lidar pipeline and component_traffic_light.launch.xml for traffic light recognition); this package hosts the perception feature launchers they combine:

launch/
├── common/                        # shared preprocess feature launchers (pointcloud downsample)
├── object_recognition/
│   ├── object_recognition_<mode>.launch.xml  # detection + tracking + prediction per sensor mode
│   ├── detection/
│   │   ├── detector/              # detectors
│   │   ├── filter/                # filters
│   │   └── merger/                # detection-side serial merger (used by the camera_lidar_radar_serial launcher)
│   └── prediction/
├── obstacle_segmentation/
├── occupancy_grid_map/
└── traffic_light_recognition/
design/module/                     # Autoware System Designer modules; same subtree as launch/
├── object_recognition/
├── obstacle_segmentation/
├── occupancy_grid_map/
└── traffic_light_recognition/

Launch files

  • object_recognition/object_recognition_<mode>.launch.xml — one feature launcher per sensor mode (camera, lidar, lidar_ptv3, lidar_radar, camera_lidar_radar, camera_lidar_radar_serial); hosts the detection stage inline together with the tracker and prediction, and pushes their module namespaces. Camera and radar input topics default to their absolute sensing paths (/sensing/camera/cameraN/..., /sensing/radar/detected_objects), so callers override only what differs. Most modes wire the multi-channel tracker to the detector outputs; camera_lidar_radar_serial instead merges the detectors in the detection stage (optionally regenerating detections through detection-by-tracker), tracks the single merged detected_objects channel, and joins radar tracked objects at the decorative tracker merger. camera runs camera-only detectors (BEV and camera VRU).
  • The clustering backend is file selection, not a flag: lidar runs rule-based euclidean clustering off the obstacle segmentation cloud, while lidar_ptv3 and lidar_radar cluster the PTv3 semantic-segmentation cloud and drop the pointcloud map filter and the object validator that only the rule-based pipeline uses. Both feed their cluster objects to the tracker on the lidar_clustering channel.
  • object_recognition/detection/{detector,filter}/ — single-purpose launchers wired together by the object-recognition launchers. merger/camera_lidar_merger.launch.xml chains the detection-side object mergers for the camera_lidar_radar_serial launcher.
  • obstacle_segmentation/ground_segmentation_<backend>.launch.xml — ground removal backends (rule_based, rule_based_cuda, ml); backend selection is file selection. ground_segmentation_{additional_lidars,ransac}.launch.xml fold extra lidars into the obstacle stream for kits that carry them.
  • occupancy_grid_map/, common/, traffic_light_recognition/ — feature launchers included directly by the system components.

Namespace and container addressing

Callers own the namespaces: the system component pushes perception and object_recognition; object_recognition_<mode>.launch.xml pushes detection/tracking/prediction. The object-recognition launchers address their internal topics through the fixed /perception/object_recognition namespace. Composable nodes load into the shared pointcloud container, whose name is exposed as pointcloud_container_name and aliased to the node/pointcloud_container argument of the detector launchers.

Config

Parameter files are resolved from a config/ tree addressed through the perception_config_pkg argument, which defaults to autoware_perception_config. A product can relocate the whole tree by setting perception_config_pkg once at its entry point — the launch configuration propagates through every include — but it must then ship the same config/ layout. Products that differ in only a few files instead override the corresponding *_param_path arguments, which every feature launcher declares with a default:

<include file="$(find-pkg-share autoware_perception_launch)/launch/object_recognition/object_recognition_lidar.launch.xml">
  <arg name="perception_config_pkg" value="my_perception_config"/>
  <arg name="FOO_NODE_param_path" value="..."/>
</include>

component_perception.launch.xml, component_traffic_light.launch.xml and planning_simulator/perception_simulation.launch.xml each declare perception_config_pkg, so a product switches the tree at the component it launches.

The launch/ tree is the interface downstream distributions adopt as-is; only package.xml, CMakeLists.txt, README.md, and changelogs stay package-local. Keep changes to this tree folder-merge compatible.

Package Dependencies

Please see <exec_depend> in package.xml.

CHANGELOG

Changelog for package autoware_perception_launch

0.53.0 (2026-10-13)

  • Merge remote-tracking branch 'origin/main' into tmp/bot/bump_version_base

  • feat(autoware_launch): run crop_box_filter standalone when agnocast is enabled (#1990)

    * feat(autoware_launch): run crop_box_filter standalone when agnocast is enabled autoware_universe#13384 moves CropBoxFilterComponent to agnocast_wrapper::Node on an agnocast-only executor, which the component containers do not set up. Start each of the eight crop_box_filter nodes as a standalone process when ENABLE_AGNOCAST=1 and keep loading them into their container otherwise. The node names, namespaces, remaps and parameters are the same in both branches.

    * refactor(autoware_launch): launch crop_box_filter through the autoware_node action ---------

  • feat(autoware_perception_launch): run scan_ground_filter standalone when agnocast is enabled (#1991)

    * feat(autoware_perception_launch): run scan_ground_filter standalone when agnocast is enabled autoware_universe#13385 moves ScanGroundFilterComponent to agnocast_wrapper::Node on an agnocast-only executor, which the component containers do not set up. Start each of the three scan_ground_filter nodes as a standalone process when ENABLE_AGNOCAST=1 and keep loading them into their container otherwise. The node names, namespaces, remaps and parameters are the same in both branches.

    * refactor(autoware_perception_launch): launch scan_ground_filter through the autoware_node action ---------

  • feat(autoware_launch): run voxel_grid_downsample_filter standalone when agnocast is enabled (#1992)

    • feat(autoware_launch): run voxel_grid_downsample_filter standalone when agnocast is enabled
    • style(pre-commit): autofix

    * refactor(autoware_launch): launch voxel_grid_downsample_filter through the autoware_node action ---------Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>

  • feat(autoware_sample_designs): compose the AD API, system and vehicle components from design modules (#1976)

    * fix(control, planning): connect the operation mode state and velocity limit inputs Wires collision_detector's operation_mode_state and remaining_distance_time_calculator's planning_velocity through the module designs; their node designs now declare the fixed topic names as remap targets.

    * fix(planning): connect the internal velocity limit path to the selector Forwards motion_velocity_planner's velocity_limit and clear_velocity_limit up to ScenarioPlanning and feeds them into external_velocity_limit_selector, whose internal inputs were unconnected.

    * fix(control, planning): connect the remaining module inputs to their sources Wires the operation mode transition manager's gate and selector inputs, the stop mode operator's vehicle status, the manual lane change handler's route availability and preferred primitive service, the parking costmap, and the route marker. Exposes the vehicle control mode service, the preferred lane service and the system hazard lights command as component interfaces.

    • fix(planning): follow the hazard lights selector input rename
    • fix(planning): add steering_status subscriber to various planning modules
    • feat(vehicle): add vehicle wrapper and launch file for vehicle interface
    • style(pre-commit): autofix
    • fix(planning): add planning factors publishers and connections for various modules
    • refactor(autoware_sample_designs): compose the AD API from design modules
    • Refactor ADAPI module structure and enhance planning factors
    • Updated AdapiRvizHelpers.module.yaml to include helper prefixes for instances and connections.
    • Modified AutowareSample.system.yaml to reorganize ADAPI components, introducing default_adapi for better structure.
    • Expanded planning factors in Planning.module.yaml, LaneDriving.module.yaml, ScenarioPlanning.module.yaml, and associated modules to include additional factors for improved behavior and motion planning.
    • Enhanced connections in behavior planning and motion planning modules to accommodate new planning factors.
    • Ensured consistency across scenario planning modules by aligning

File truncated at 100 lines see the full file

Package Dependencies

Deps Name
ament_cmake_auto
autoware_cmake
autoware_agnocast_wrapper
autoware_bevfusion
autoware_bytetrack
autoware_compare_map_segmentation
autoware_crosswalk_traffic_light_estimator
autoware_cuda_pointcloud_preprocessor
autoware_detected_object_feature_remover
autoware_detected_object_validation
autoware_detection_by_tracker
autoware_elevation_map_loader
autoware_euclidean_cluster
autoware_ground_segmentation
autoware_ground_segmentation_cuda
autoware_image_object_locator
autoware_image_projection_based_fusion
autoware_image_transport_decompressor
autoware_lidar_apollo_instance_segmentation
autoware_lidar_centerpoint
autoware_lidar_transfusion
autoware_map_based_prediction
autoware_multi_object_tracker
autoware_object_merger
autoware_object_sorter
autoware_occupancy_grid_map_outlier_filter
autoware_perception_config
autoware_pointcloud_preprocessor
autoware_probabilistic_occupancy_grid_map
autoware_ptv3
autoware_raindrop_cluster_filter
autoware_shape_estimation
autoware_simpl_prediction
autoware_simple_object_merger
autoware_tensorrt_bevdet
autoware_tensorrt_yolox
autoware_tracking_object_merger
autoware_traffic_light_arbiter
autoware_traffic_light_category_merger
autoware_traffic_light_classifier
autoware_traffic_light_fine_detector
autoware_traffic_light_map_based_detector
autoware_traffic_light_multi_camera_fusion
autoware_traffic_light_occlusion_predictor
autoware_traffic_light_selector
autoware_traffic_light_visualization
rclcpp_components
topic_tools
ament_lint_auto
autoware_lint_common

System Dependencies

No direct system dependencies.

Dependant Packages

Name Deps
autoware_launch
reaction_analyzer

Launch files

  • launch/common/pointcloud_downsample.launch.xml
      • pointcloud_container_name
      • cuda_pointcloud_preprocessing [default: false]
      • input/pointcloud
      • output/pointcloud
      • common_downsample_voxel_size_x [default: 0.05]
      • common_downsample_voxel_size_y [default: 0.05]
      • common_downsample_voxel_size_z [default: 0.05]
  • launch/object_recognition/detection/detector/camera_2d_detector.launch.xml
      • image_raw0 [default: /sensing/camera/camera0/image_raw]
      • image_raw1 [default: /sensing/camera/camera1/image_raw]
      • image_raw2 [default: /sensing/camera/camera2/image_raw]
      • image_raw3 [default: /sensing/camera/camera3/image_raw]
      • image_raw4 [default: /sensing/camera/camera4/image_raw]
      • image_raw5 [default: /sensing/camera/camera5/image_raw]
      • image_raw6 [default: /sensing/camera/camera6/image_raw]
      • image_raw7 [default: /sensing/camera/camera7/image_raw]
      • image_raw8 [default: /sensing/camera/camera8/image_raw]
      • image_raw9 [default: /sensing/camera/camera9/image_raw]
      • image_number [default: 1]
      • camera_index [default: 0]
      • use_bytetrack [default: true]
      • enable_visualizer [default: false]
      • camera_2d_detector/model_path
      • camera_2d_detector/label_path
      • camera_2d_detector/color_map_path
      • camera_2d_detector/roi_remap_path
      • camera_2d_detector/roi_to_semseg_remap_path
      • tensorrt_yolox_ns [default: ]
  • launch/object_recognition/detection/detector/camera_bev_detector.launch.xml
      • input/camera0/image
      • input/camera0/info
      • input/camera1/image
      • input/camera1/info
      • input/camera2/image
      • input/camera2/info
      • input/camera3/image
      • input/camera3/info
      • input/camera4/image
      • input/camera4/info
      • input/camera5/image
      • input/camera5/info
      • output/objects
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • bevdet_model_name [default: bevdet_one_lt_d]
      • bevdet_model_path [default: $(var data_path)/tensorrt_bevdet]
  • launch/object_recognition/detection/detector/camera_lidar_detector.launch.xml
      • ns
      • lidar_detection_model_type
      • lidar_detection_model_name
      • use_low_intensity_cluster_filter
      • use_image_segmentation_based_filter
      • number_of_cameras
      • input/camera0/image
      • input/camera0/info
      • input/camera0/rois
      • input/camera1/image
      • input/camera1/info
      • input/camera1/rois
      • input/camera2/image
      • input/camera2/info
      • input/camera2/rois
      • input/camera3/image
      • input/camera3/info
      • input/camera3/rois
      • input/camera4/image
      • input/camera4/info
      • input/camera4/rois
      • input/camera5/image
      • input/camera5/info
      • input/camera5/rois
      • input/camera6/image
      • input/camera6/info
      • input/camera6/rois
      • input/camera7/image
      • input/camera7/info
      • input/camera7/rois
      • input/camera8/image
      • input/camera8/info
      • input/camera8/rois
      • input/concatenation_info
      • segmentation_pointcloud_fusion_camera_ids
      • image_topic_name
      • sync_param_path
      • voxel_grid_based_euclidean_param_path
      • lidar_model_param_path
      • pointpainting_model_path [default: $(var data_path)/image_projection_based_fusion]
      • pointpainting_fusion_common_param_path
      • roi_cluster_fusion_param_path
      • node/pointcloud_container
      • input/pointcloud
      • input/pointcloud_map/pointcloud
      • input/obstacle_segmentation/pointcloud
      • output/ml_detector/objects
      • output/rule_detector/objects
      • camera_2d_detector/model_path
      • camera_2d_detector/label_path
      • camera_2d_detector/color_map_path
      • camera_2d_detector/roi_remap_path
      • camera_2d_detector/roi_to_semseg_remap_path
      • enable_2d_detection [default: false]
  • launch/object_recognition/detection/detector/camera_lidar_irregular_object_detector.launch.xml
      • pipeline_ns
      • node/pointcloud_container
      • output/objects [default: objects]
      • input/concatenation_info
      • input/pointcloud
      • irregular_object_crop_box_filter_node_param_path
      • irregular_object_scan_ground_filter_node_param_path
      • irregular_object_roi_pointcloud_fusion_node_param_path
      • irregular_object_fusion_camera_param_path
      • sync_param_path
  • launch/object_recognition/detection/detector/camera_vru_detector.launch.xml
      • input/camera0/info [default: /sensing/camera/camera0/camera_info]
      • input/camera0/rois [default: /perception/object_recognition/detection/rois0]
      • input/camera1/info [default: /sensing/camera/camera1/camera_info]
      • input/camera1/rois [default: /perception/object_recognition/detection/rois1]
      • input/camera2/info [default: /sensing/camera/camera2/camera_info]
      • input/camera2/rois [default: /perception/object_recognition/detection/rois2]
      • input/camera3/info [default: /sensing/camera/camera3/camera_info]
      • input/camera3/rois [default: /perception/object_recognition/detection/rois3]
      • input/camera4/info [default: /sensing/camera/camera4/camera_info]
      • input/camera4/rois [default: /perception/object_recognition/detection/rois4]
      • input/camera5/info [default: /sensing/camera/camera5/camera_info]
      • input/camera5/rois [default: /perception/object_recognition/detection/rois5]
      • input/camera6/info [default: /sensing/camera/camera6/camera_info]
      • input/camera6/rois [default: /perception/object_recognition/detection/rois6]
      • input/camera7/info [default: /sensing/camera/camera7/camera_info]
      • input/camera7/rois [default: /perception/object_recognition/detection/rois7]
      • output/objects [default: /perception/object_recognition/detection/camera_vru/objects]
      • bbox_object_locator_param_path
      • rois_ids [default: [0, 1]]
  • launch/object_recognition/detection/detector/irregular_object_detector.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • target_container
      • base_frame [default: base_link]
      • use_intra_process [default: true]
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • input/concatenation_info [default: /sensing/lidar/concatenated/pointcloud_info]
      • output/clusters
      • crop_box_filter_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/crop_box_filter.param.yaml]
      • scan_ground_filter_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/scan_ground_filter.param.yaml]
      • roi_pointcloud_fusion_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/roi_pointcloud_fusion.param.yaml]
      • fusion_camera_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/fusion_cameras.param.yaml]
      • sync_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/fusion_common.param.yaml]
  • launch/object_recognition/detection/detector/lidar_dnn_detector.launch.xml
      • lidar_detection_model_type
      • lidar_detection_model_name
      • lidar_model_param_path
      • bevfusion_model_path [default: $(var data_path)/bevfusion]
      • centerpoint_model_path [default: $(var data_path)/lidar_centerpoint]
      • centerpoint_common_param_path [default: $(find-pkg-share autoware_lidar_centerpoint)/config/centerpoint_common.param.yaml]
      • transfusion_model_path [default: $(var data_path)/lidar_transfusion]
      • use_short_range_detection [default: false]
      • lidar_short_range_detection_model_type
      • node/pointcloud_container
      • input/pointcloud
      • output/objects
      • output/short_range_objects
  • launch/object_recognition/detection/detector/lidar_rule_detector.launch.xml
      • ns
      • node/pointcloud_container
      • input/pointcloud_map/pointcloud
      • input/obstacle_segmentation/pointcloud
      • output/cluster_objects
      • output/objects
      • voxel_grid_based_euclidean_param_path
      • use_low_height_cropbox [default: false]
  • launch/object_recognition/detection/detector/lidar_semantic_segmentation_detector.launch.xml
      • euclidean_clustering/param_path
      • input/pointcloud
      • output/objects
  • launch/object_recognition/detection/detector/tracker_based_detector.launch.xml
      • detection_by_tracker_param_path
      • input/clusters
      • input/tracked_objects
      • output/objects
  • launch/object_recognition/detection/filter/object_filter.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • objects_filter_method [default: lanelet_filter]
      • object_recognition_detection_object_lanelet_filter_param_path [default: $(var perception_config_dir)/object_recognition/detection/object_filter/object_lanelet_filter.param.yaml]
      • object_recognition_detection_object_position_filter_param_path [default: $(var perception_config_dir)/object_recognition/detection/object_filter/object_position_filter.param.yaml]
      • input/objects
      • output/objects
  • launch/object_recognition/detection/filter/object_validator.launch.xml
      • objects_validation_method
      • object_recognition_detection_obstacle_pointcloud_based_validator_param_path
      • input/obstacle_pointcloud
      • input/objects
      • output/objects
  • launch/object_recognition/detection/filter/pointcloud_downsample_filter.launch.xml
      • target_container
      • use_intra_process [default: true]
      • input/pointcloud
      • output/pointcloud
      • downsample_voxel_size [default: 0.1]
  • launch/object_recognition/detection/filter/pointcloud_map_filter.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • target_container
      • use_intra_process [default: true]
      • input/pointcloud
      • output/pointcloud
      • output/downsampled/pointcloud [default: /perception/obstacle_segmentation/pointcloud_map_filtered/downsampled/pointcloud]
      • downsample_voxel_size [default: 0.1]
      • voxel_based_compare_map_filter_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/pointcloud_filter/voxel_based_compare_map_filter.param.yaml]
  • launch/object_recognition/detection/filter/radar_filter.launch.xml
      • object_sorter_param_path
      • radar_lanelet_filtering_range_param_path
      • input/radar
      • output/objects
  • launch/object_recognition/detection/merger/camera_lidar_merger.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • object_recognition_detection_object_merger_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/detection/object_merger/data_association_matrix.param.yaml]
      • object_recognition_detection_object_merger_distance_threshold_list_path [default: $(var perception_config_dir)/object_recognition/detection/object_merger/overlapped_judge.param.yaml]
      • ml_camera_lidar_object_association_merger_param_path [default: $(var perception_config_dir)/object_recognition/detection/object_merger/ml_camera_lidar_object_association_merger.param.yaml]
      • roi_detected_object_fusion_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/roi_detected_object_fusion.param.yaml]
      • sync_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/fusion_common.param.yaml]
      • lidar_detection_model_type
      • use_detection_by_tracker
      • use_irregular_object_detector
      • use_roi_detected_object_validator
      • euclidean_clustering/shape_policy
      • use_object_filter
      • objects_filter_method
      • number_of_cameras
      • input/camera0/image
      • input/camera0/info
      • input/camera0/rois
      • input/camera1/image
      • input/camera1/info
      • input/camera1/rois
      • input/camera2/image
      • input/camera2/info
      • input/camera2/rois
      • input/camera3/image
      • input/camera3/info
      • input/camera3/rois
      • input/camera4/image
      • input/camera4/info
      • input/camera4/rois
      • input/camera5/image
      • input/camera5/info
      • input/camera5/rois
      • input/camera6/image
      • input/camera6/info
      • input/camera6/rois
      • input/camera7/image
      • input/camera7/info
      • input/camera7/rois
      • input/camera8/image
      • input/camera8/info
      • input/camera8/rois
      • input/concatenation_info [default: /sensing/lidar/concatenated/pointcloud_info]
      • input/lidar_ml/objects
      • input/lidar_cluster/objects
      • input/irregular_object/objects [default: ]
      • input/detection_by_tracker/objects
      • output/objects [default: objects]
      • alpha_merger_priority_mode [default: 0]
  • launch/object_recognition/object_recognition_camera.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • use_camera_vru_detector [default: false]
      • camera_vru_detector_rois_ids [default: [0]]
      • use_camera_bev_detector [default: true]
      • use_vector_map [default: true]
      • tracker_publish_merged_objects [default: false]
      • input/camera0/image [default: /sensing/camera/camera0/image_rect_color]
      • input/camera0/info [default: /sensing/camera/camera0/camera_info]
      • input/camera0/rois [default: /perception/object_recognition/detection/rois0]
      • input/camera1/image [default: /sensing/camera/camera1/image_rect_color]
      • input/camera1/info [default: /sensing/camera/camera1/camera_info]
      • input/camera1/rois [default: /perception/object_recognition/detection/rois1]
      • input/camera2/image [default: /sensing/camera/camera2/image_rect_color]
      • input/camera2/info [default: /sensing/camera/camera2/camera_info]
      • input/camera2/rois [default: /perception/object_recognition/detection/rois2]
      • input/camera3/image [default: /sensing/camera/camera3/image_rect_color]
      • input/camera3/info [default: /sensing/camera/camera3/camera_info]
      • input/camera3/rois [default: /perception/object_recognition/detection/rois3]
      • input/camera4/image [default: /sensing/camera/camera4/image_rect_color]
      • input/camera4/info [default: /sensing/camera/camera4/camera_info]
      • input/camera4/rois [default: /perception/object_recognition/detection/rois4]
      • input/camera5/image [default: /sensing/camera/camera5/image_rect_color]
      • input/camera5/info [default: /sensing/camera/camera5/camera_info]
      • input/camera5/rois [default: /perception/object_recognition/detection/rois5]
      • input/camera6/image [default: /sensing/camera/camera6/image_rect_color]
      • input/camera6/info [default: /sensing/camera/camera6/camera_info]
      • input/camera6/rois [default: /perception/object_recognition/detection/rois6]
      • input/camera7/image [default: /sensing/camera/camera7/image_rect_color]
      • input/camera7/info [default: /sensing/camera/camera7/camera_info]
      • input/camera7/rois [default: /perception/object_recognition/detection/rois7]
      • output/tracker_merged_objects [default: /perception/object_recognition/detection/objects]
      • output/tracked_objects [default: /perception/object_recognition/tracking/objects]
      • output/predicted_objects [default: /perception/object_recognition/objects]
      • near_range_camera_vru_param_path [default: $(var perception_config_dir)/object_recognition/detection/camera_vru_detection/near_range_camera_vru_detector.param.yaml]
      • object_recognition_tracking_multi_object_tracker_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/data_association_matrix.param.yaml]
      • object_recognition_tracking_multi_object_tracker_input_channels_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/input_channels.param.yaml]
      • object_recognition_tracking_multi_object_tracker_node_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/multi_object_tracker_node.param.yaml]
  • launch/object_recognition/object_recognition_camera_lidar_radar.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • lidar_detection_model [default: centerpoint]
      • use_short_range_detection [default: false]
      • lidar_short_range_detection_model_type [default: centerpoint_short_range]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • bevfusion_model_path [default: $(var data_path)/bevfusion]
      • pointpainting_model_path [default: $(var data_path)/image_projection_based_fusion]
      • camera_2d_detector/model_path [default: $(var data_path)/tensorrt_yolox/yolox-sPlus-opt-pseudoV2-T4-960x960-T4-seg16cls.onnx]
      • camera_2d_detector/label_path [default: $(var data_path)/tensorrt_yolox/label.txt]
      • camera_2d_detector/color_map_path [default: $(var data_path)/tensorrt_yolox/semseg_color_map.csv]
      • camera_2d_detector/roi_remap_path [default: $(var perception_config_dir)/traffic_light_recognition/tensorrt_yolox/roi_label_remap.csv]
      • camera_2d_detector/roi_to_semseg_remap_path [default: $(var perception_config_dir)/traffic_light_recognition/tensorrt_yolox/roi_to_semseg_label_remap.csv]
      • use_semantic_segmentation_ptv3 [default: false]
      • use_object_filter [default: true]
      • use_pointcloud_map [default: true]
      • use_validator [default: false]
      • objects_validation_method [default: obstacle_pointcloud]
      • use_low_height_cropbox [default: false]
      • use_low_intensity_cluster_filter [default: true]
      • use_image_segmentation_based_filter [default: false]
      • use_irregular_object_detector [default: true]
      • use_camera_vru_detector [default: false]
      • camera_vru_detector_rois_ids [default: [0]]
      • use_camera_only_detection [default: false]
      • use_vector_map [default: true]
      • tracker_publish_merged_objects [default: false]
      • number_of_cameras [default: 6]
      • pointcloud_container_name [default: pointcloud_container]
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • input/semantic_segmentation/pointcloud [default: /perception/segmented/pointcloud]
      • input/obstacle_segmentation/pointcloud [default: /perception/obstacle_segmentation/pointcloud]
      • input/camera0/image [default: /sensing/camera/camera0/image_rect_color]
      • input/camera0/info [default: /sensing/camera/camera0/camera_info]
      • input/camera0/rois [default: /perception/object_recognition/detection/rois0]
      • input/camera1/image [default: /sensing/camera/camera1/image_rect_color]
      • input/camera1/info [default: /sensing/camera/camera1/camera_info]
      • input/camera1/rois [default: /perception/object_recognition/detection/rois1]
      • input/camera2/image [default: /sensing/camera/camera2/image_rect_color]
      • input/camera2/info [default: /sensing/camera/camera2/camera_info]
      • input/camera2/rois [default: /perception/object_recognition/detection/rois2]
      • input/camera3/image [default: /sensing/camera/camera3/image_rect_color]
      • input/camera3/info [default: /sensing/camera/camera3/camera_info]
      • input/camera3/rois [default: /perception/object_recognition/detection/rois3]
      • input/camera4/image [default: /sensing/camera/camera4/image_rect_color]
      • input/camera4/info [default: /sensing/camera/camera4/camera_info]
      • input/camera4/rois [default: /perception/object_recognition/detection/rois4]
      • input/camera5/image [default: /sensing/camera/camera5/image_rect_color]
      • input/camera5/info [default: /sensing/camera/camera5/camera_info]
      • input/camera5/rois [default: /perception/object_recognition/detection/rois5]
      • input/camera6/image [default: /sensing/camera/camera6/image_rect_color]
      • input/camera6/info [default: /sensing/camera/camera6/camera_info]
      • input/camera6/rois [default: /perception/object_recognition/detection/rois6]
      • input/camera7/image [default: /sensing/camera/camera7/image_rect_color]
      • input/camera7/info [default: /sensing/camera/camera7/camera_info]
      • input/camera7/rois [default: /perception/object_recognition/detection/rois7]
      • input/camera8/image [default: /sensing/camera/camera8/image_rect_color]
      • input/camera8/info [default: /sensing/camera/camera8/camera_info]
      • input/camera8/rois [default: /perception/object_recognition/detection/rois8]
      • input/concatenation_info [default: /sensing/lidar/concatenated/pointcloud_info]
      • input/radar [default: /sensing/radar/detected_objects]
      • input/camera_only/objects [default: /perception/object_recognition/detection/camera_only/objects]
      • input/camera_only/channel [default: camera_streampetr]
      • image_topic_name [default: image_rect_color]
      • segmentation_pointcloud_fusion_camera_ids [default: [0,2,4]]
      • output/tracker_merged_objects [default: /perception/object_recognition/detection/objects]
      • output/tracked_objects [default: /perception/object_recognition/tracking/objects]
      • output/predicted_objects [default: /perception/object_recognition/objects]
      • sync_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/fusion_common.param.yaml]
      • roi_cluster_fusion_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/roi_cluster_fusion.param.yaml]
      • pointpainting_fusion_common_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/pointpainting_common.param.yaml]
      • irregular_object_crop_box_filter_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/crop_box_filter.param.yaml]
      • irregular_object_scan_ground_filter_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/scan_ground_filter.param.yaml]
      • irregular_object_roi_pointcloud_fusion_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/roi_pointcloud_fusion.param.yaml]
      • irregular_object_fusion_camera_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/fusion_cameras.param.yaml]
      • lidar_model_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model]
      • centerpoint_common_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model/centerpoint_common.param.yaml]
      • voxel_grid_based_euclidean_param_path [default: $(var perception_config_dir)/object_recognition/detection/clustering/voxel_grid_based_euclidean_cluster.param.yaml]
      • radar_lanelet_filtering_range_param_path [default: $(var perception_config_dir)/object_recognition/detection/object_filter/radar_lanelet_filter.param.yaml]
      • object_recognition_detection_object_sorter_radar_param_path [default: $(var perception_config_dir)/object_recognition/detection/radar_filter/detected_object_sorter.param.yaml]
      • near_range_camera_vru_param_path [default: $(var perception_config_dir)/object_recognition/detection/camera_vru_detection/near_range_camera_vru_detector.param.yaml]
      • object_recognition_detection_obstacle_pointcloud_based_validator_param_path [default: $(var perception_config_dir)/object_recognition/detection/detected_object_validation/obstacle_pointcloud_based_validator.param.yaml]
      • object_recognition_tracking_multi_object_tracker_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/data_association_matrix.param.yaml]
      • object_recognition_tracking_multi_object_tracker_input_channels_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/input_channels.param.yaml]
      • object_recognition_tracking_multi_object_tracker_node_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/multi_object_tracker_node.param.yaml]
      • euclidean_clustering/param_path [default: $(var perception_config_dir)/object_recognition/detection/euclidean_cluster/label_based_euclidean_cluster.param.yaml]
  • launch/object_recognition/object_recognition_camera_lidar_radar_serial.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • lidar_detection_model [default: centerpoint]
      • use_short_range_detection [default: false]
      • lidar_short_range_detection_model_type [default: centerpoint_short_range]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • bevfusion_model_path [default: $(var data_path)/bevfusion]
      • pointpainting_model_path [default: $(var data_path)/image_projection_based_fusion]
      • camera_2d_detector/model_path [default: $(var data_path)/tensorrt_yolox/yolox-sPlus-opt-pseudoV2-T4-960x960-T4-seg16cls.onnx]
      • camera_2d_detector/label_path [default: $(var data_path)/tensorrt_yolox/label.txt]
      • camera_2d_detector/color_map_path [default: $(var data_path)/tensorrt_yolox/semseg_color_map.csv]
      • camera_2d_detector/roi_remap_path [default: $(var perception_config_dir)/traffic_light_recognition/tensorrt_yolox/roi_label_remap.csv]
      • camera_2d_detector/roi_to_semseg_remap_path [default: $(var perception_config_dir)/traffic_light_recognition/tensorrt_yolox/roi_to_semseg_label_remap.csv]
      • use_object_filter [default: true]
      • objects_filter_method [default: lanelet_filter]
      • use_pointcloud_map [default: true]
      • use_validator [default: false]
      • objects_validation_method [default: obstacle_pointcloud]
      • use_roi_detected_object_validator [default: true]
      • use_detection_by_tracker [default: true]
      • use_low_height_cropbox [default: false]
      • use_low_intensity_cluster_filter [default: true]
      • use_image_segmentation_based_filter [default: false]
      • use_irregular_object_detector [default: true]
      • ml_camera_lidar_merger_priority_mode [default: 0]
      • use_vector_map [default: true]
      • number_of_cameras [default: 6]
      • pointcloud_container_name [default: pointcloud_container]
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • input/obstacle_segmentation/pointcloud [default: /perception/obstacle_segmentation/pointcloud]
      • input/camera0/image [default: /sensing/camera/camera0/image_rect_color]
      • input/camera0/info [default: /sensing/camera/camera0/camera_info]
      • input/camera0/rois [default: /perception/object_recognition/detection/rois0]
      • input/camera1/image [default: /sensing/camera/camera1/image_rect_color]
      • input/camera1/info [default: /sensing/camera/camera1/camera_info]
      • input/camera1/rois [default: /perception/object_recognition/detection/rois1]
      • input/camera2/image [default: /sensing/camera/camera2/image_rect_color]
      • input/camera2/info [default: /sensing/camera/camera2/camera_info]
      • input/camera2/rois [default: /perception/object_recognition/detection/rois2]
      • input/camera3/image [default: /sensing/camera/camera3/image_rect_color]
      • input/camera3/info [default: /sensing/camera/camera3/camera_info]
      • input/camera3/rois [default: /perception/object_recognition/detection/rois3]
      • input/camera4/image [default: /sensing/camera/camera4/image_rect_color]
      • input/camera4/info [default: /sensing/camera/camera4/camera_info]
      • input/camera4/rois [default: /perception/object_recognition/detection/rois4]
      • input/camera5/image [default: /sensing/camera/camera5/image_rect_color]
      • input/camera5/info [default: /sensing/camera/camera5/camera_info]
      • input/camera5/rois [default: /perception/object_recognition/detection/rois5]
      • input/camera6/image [default: /sensing/camera/camera6/image_rect_color]
      • input/camera6/info [default: /sensing/camera/camera6/camera_info]
      • input/camera6/rois [default: /perception/object_recognition/detection/rois6]
      • input/camera7/image [default: /sensing/camera/camera7/image_rect_color]
      • input/camera7/info [default: /sensing/camera/camera7/camera_info]
      • input/camera7/rois [default: /perception/object_recognition/detection/rois7]
      • input/camera8/image [default: /sensing/camera/camera8/image_rect_color]
      • input/camera8/info [default: /sensing/camera/camera8/camera_info]
      • input/camera8/rois [default: /perception/object_recognition/detection/rois8]
      • input/concatenation_info [default: /sensing/lidar/concatenated/pointcloud_info]
      • input/radar/tracked_objects [default: /sensing/radar/tracked_objects]
      • image_topic_name [default: image_rect_color]
      • segmentation_pointcloud_fusion_camera_ids [default: [0,2,4]]
      • output/tracked_objects [default: /perception/object_recognition/tracking/objects]
      • output/predicted_objects [default: /perception/object_recognition/objects]
      • sync_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/fusion_common.param.yaml]
      • roi_cluster_fusion_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/roi_cluster_fusion.param.yaml]
      • pointpainting_fusion_common_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/pointpainting_common.param.yaml]
      • irregular_object_crop_box_filter_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/crop_box_filter.param.yaml]
      • irregular_object_scan_ground_filter_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/scan_ground_filter.param.yaml]
      • irregular_object_roi_pointcloud_fusion_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/roi_pointcloud_fusion.param.yaml]
      • irregular_object_fusion_camera_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/fusion_cameras.param.yaml]
      • lidar_model_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model]
      • centerpoint_common_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model/centerpoint_common.param.yaml]
      • voxel_grid_based_euclidean_param_path [default: $(var perception_config_dir)/object_recognition/detection/clustering/voxel_grid_based_euclidean_cluster.param.yaml]
      • detection_by_tracker_param_path [default: $(var perception_config_dir)/object_recognition/detection/detection_by_tracker/detection_by_tracker.param.yaml]
      • object_recognition_detection_obstacle_pointcloud_based_validator_param_path [default: $(var perception_config_dir)/object_recognition/detection/detected_object_validation/obstacle_pointcloud_based_validator.param.yaml]
      • object_recognition_detection_object_merger_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/detection/object_merger/data_association_matrix.param.yaml]
      • object_recognition_detection_object_merger_distance_threshold_list_path [default: $(var perception_config_dir)/object_recognition/detection/object_merger/overlapped_judge.param.yaml]
      • ml_camera_lidar_object_association_merger_param_path [default: $(var perception_config_dir)/object_recognition/detection/object_merger/ml_camera_lidar_object_association_merger.param.yaml]
      • roi_detected_object_fusion_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/roi_detected_object_fusion.param.yaml]
      • object_recognition_tracking_multi_object_tracker_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/data_association_matrix.param.yaml]
      • object_recognition_tracking_multi_object_tracker_input_channels_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/input_channels.param.yaml]
      • object_recognition_tracking_multi_object_tracker_node_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/multi_object_tracker_node.param.yaml]
      • object_recognition_tracking_radar_tracked_object_sorter_param_path [default: $(var perception_config_dir)/object_recognition/tracking/radar_object_tracker/tracked_object_sorter.param.yaml]
      • object_recognition_tracking_radar_tracked_object_lanelet_filter_param_path [default: $(var perception_config_dir)/object_recognition/tracking/radar_object_tracker/tracked_object_lanelet_filter.param.yaml]
      • object_recognition_tracking_object_merger_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/tracking/tracking_object_merger/data_association_matrix.param.yaml]
      • object_recognition_tracking_object_merger_node_param_path [default: $(var perception_config_dir)/object_recognition/tracking/tracking_object_merger/decorative_tracker_merger.param.yaml]
      • euclidean_clustering/shape_policy [default: 0]
  • launch/object_recognition/object_recognition_lidar.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • lidar_detection_model [default: centerpoint]
      • use_short_range_detection [default: false]
      • lidar_short_range_detection_model_type [default: centerpoint_short_range]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • bevfusion_model_path [default: $(var data_path)/bevfusion]
      • use_pointcloud_map [default: true]
      • use_validator [default: false]
      • objects_validation_method [default: obstacle_pointcloud]
      • use_object_filter [default: true]
      • objects_filter_method [default: lanelet_filter]
      • use_low_height_cropbox [default: false]
      • use_vector_map [default: true]
      • tracker_publish_merged_objects [default: false]
      • pointcloud_container_name [default: pointcloud_container]
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • input/obstacle_segmentation/pointcloud [default: /perception/obstacle_segmentation/pointcloud]
      • output/tracker_merged_objects [default: /perception/object_recognition/detection/objects]
      • output/tracked_objects [default: /perception/object_recognition/tracking/objects]
      • output/predicted_objects [default: /perception/object_recognition/objects]
      • lidar_model_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model]
      • centerpoint_common_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model/centerpoint_common.param.yaml]
      • voxel_grid_based_euclidean_param_path [default: $(var perception_config_dir)/object_recognition/detection/clustering/voxel_grid_based_euclidean_cluster.param.yaml]
      • object_recognition_detection_obstacle_pointcloud_based_validator_param_path [default: $(var perception_config_dir)/object_recognition/detection/detected_object_validation/obstacle_pointcloud_based_validator.param.yaml]
      • object_recognition_tracking_multi_object_tracker_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/data_association_matrix.param.yaml]
      • object_recognition_tracking_multi_object_tracker_input_channels_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/input_channels.param.yaml]
      • object_recognition_tracking_multi_object_tracker_node_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/multi_object_tracker_node.param.yaml]
  • launch/object_recognition/object_recognition_lidar_ptv3.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • lidar_detection_model [default: centerpoint]
      • use_short_range_detection [default: false]
      • lidar_short_range_detection_model_type [default: centerpoint_short_range]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • bevfusion_model_path [default: $(var data_path)/bevfusion]
      • use_vector_map [default: true]
      • tracker_publish_merged_objects [default: false]
      • pointcloud_container_name [default: pointcloud_container]
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • input/semantic_segmentation/pointcloud [default: /perception/segmented/pointcloud]
      • output/tracker_merged_objects [default: /perception/object_recognition/detection/objects]
      • output/tracked_objects [default: /perception/object_recognition/tracking/objects]
      • output/predicted_objects [default: /perception/object_recognition/objects]
      • lidar_model_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model]
      • centerpoint_common_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model/centerpoint_common.param.yaml]
      • object_recognition_tracking_multi_object_tracker_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/data_association_matrix.param.yaml]
      • object_recognition_tracking_multi_object_tracker_input_channels_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/input_channels.param.yaml]
      • object_recognition_tracking_multi_object_tracker_node_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/multi_object_tracker_node.param.yaml]
      • euclidean_clustering/param_path [default: $(var perception_config_dir)/object_recognition/detection/euclidean_cluster/label_based_euclidean_cluster.param.yaml]
  • launch/object_recognition/object_recognition_lidar_radar.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • lidar_detection_model [default: centerpoint]
      • use_short_range_detection [default: false]
      • lidar_short_range_detection_model_type [default: centerpoint_short_range]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • bevfusion_model_path [default: $(var data_path)/bevfusion]
      • use_low_height_cropbox [default: false]
      • use_vector_map [default: true]
      • tracker_publish_merged_objects [default: false]
      • pointcloud_container_name [default: pointcloud_container]
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • input/semantic_segmentation/pointcloud [default: /perception/segmented/pointcloud]
      • input/radar [default: /sensing/radar/detected_objects]
      • output/tracker_merged_objects [default: /perception/object_recognition/detection/objects]
      • output/tracked_objects [default: /perception/object_recognition/tracking/objects]
      • output/predicted_objects [default: /perception/object_recognition/objects]
      • lidar_model_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model]
      • centerpoint_common_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model/centerpoint_common.param.yaml]
      • voxel_grid_based_euclidean_param_path [default: $(var perception_config_dir)/object_recognition/detection/clustering/voxel_grid_based_euclidean_cluster.param.yaml]
      • radar_lanelet_filtering_range_param_path [default: $(var perception_config_dir)/object_recognition/detection/object_filter/radar_lanelet_filter.param.yaml]
      • object_recognition_detection_object_sorter_radar_param_path [default: $(var perception_config_dir)/object_recognition/detection/radar_filter/detected_object_sorter.param.yaml]
      • object_recognition_tracking_multi_object_tracker_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/data_association_matrix.param.yaml]
      • object_recognition_tracking_multi_object_tracker_input_channels_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/input_channels.param.yaml]
      • object_recognition_tracking_multi_object_tracker_node_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/multi_object_tracker_node.param.yaml]
      • euclidean_clustering/param_path [default: $(var perception_config_dir)/object_recognition/detection/euclidean_cluster/label_based_euclidean_cluster.param.yaml]
  • launch/object_recognition/prediction/prediction.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • use_vector_map [default: false]
      • prediction_model_type [default: map_based]
      • input/objects [default: /perception/object_recognition/tracking/objects]
      • output/objects [default: /perception/object_recognition/objects]
      • object_recognition_prediction_map_based_prediction_param_path [default: $(var perception_config_dir)/object_recognition/prediction/map_based_prediction.param.yaml]
  • launch/obstacle_segmentation/ground_segmentation_additional_lidars.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • target_container
      • lidar_name
      • base_frame [default: base_link]
      • use_intra_process [default: true]
      • input/main/pointcloud
      • input/lidar/pointcloud [default: /sensing/lidar/$(var lidar_name)/pointcloud]
      • output/pointcloud
      • output/pointcloud_info [default: $(var output/pointcloud)_info]
      • crop_box_margin_max_z [default: 2.5]
      • crop_box_margin_min_z [default: -2.5]
      • crop_box_filter_node_param_path [default: $(var perception_config_dir)/obstacle_segmentation/ground_segmentation/crop_box_filter.param.yaml]
      • scan_ground_filter_node_param_path [default: $(var perception_config_dir)/obstacle_segmentation/ground_segmentation/scan_ground_filter.param.yaml]
  • launch/obstacle_segmentation/ground_segmentation_ml.launch.xml
      • pointcloud_container_name [default: pointcloud_container]
      • use_pointcloud_container [default: true]
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • output/pointcloud [default: /perception/obstacle_segmentation/pointcloud]
      • output/pointcloud/segmentation [default: /perception/segmented/pointcloud]
      • output/pointcloud/visualization [default: /perception/segmented/pointcloud/visualization]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • model_variant [default: ptv3]
      • model_path [default: $(var data_path)/ptv3]
  • launch/obstacle_segmentation/ground_segmentation_ransac.launch.xml
      • target_container
      • base_frame [default: base_link]
      • use_intra_process [default: true]
      • input_topics
      • input/main/pointcloud
      • output/pointcloud
      • output/pointcloud_info [default: $(var output/pointcloud)_info]
      • detection_area_crop_box_filter_node_param_path
      • ransac_ground_filter_node_param_path [default: $(find-pkg-share autoware_ground_segmentation)/config/ransac_ground_filter.param.yaml]
  • launch/obstacle_segmentation/ground_segmentation_rule_based.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • target_container
      • base_frame [default: base_link]
      • use_intra_process [default: true]
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • output/pointcloud [default: /perception/obstacle_segmentation/pointcloud]
      • crop_box_margin_max_z [default: 2.5]
      • crop_box_margin_min_z [default: -2.5]
      • crop_box_filter_node_param_path [default: $(var perception_config_dir)/obstacle_segmentation/ground_segmentation/crop_box_filter.param.yaml]
      • scan_ground_filter_node_param_path [default: $(var perception_config_dir)/obstacle_segmentation/ground_segmentation/scan_ground_filter.param.yaml]
  • launch/obstacle_segmentation/ground_segmentation_rule_based_cuda.launch.xml
      • target_container
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • output/pointcloud [default: /perception/obstacle_segmentation/pointcloud]
      • output/ground_pointcloud [default: /perception/obstacle_segmentation/ground_pointcloud]
      • cuda_scan_ground_segmentation_filter_node_param_path [default: $(find-pkg-share autoware_ground_segmentation_cuda)/config/cuda_scan_ground_segmentation_filter.param.yaml]
  • launch/occupancy_grid_map/probabilistic_occupancy_grid_map.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • input/obstacle_pointcloud [default: concatenated/pointcloud]
      • input/raw_pointcloud [default: no_ground/oneshot/pointcloud]
      • output [default: /perception/occupancy_grid_map/map]
      • use_intra_process [default: false]
      • use_multithread [default: false]
      • pointcloud_container_name [default: pointcloud_container]
      • occupancy_grid_map_method
      • occupancy_grid_map_param_path [default: $(var perception_config_dir)/occupancy_grid_map/$(var occupancy_grid_map_method).param.yaml]
      • occupancy_grid_map_updater
      • occupancy_grid_map_updater_param_path [default: $(var perception_config_dir)/occupancy_grid_map/$(var occupancy_grid_map_updater)_updater.param.yaml]
      • input_obstacle_pointcloud [default: false]
      • input_obstacle_and_raw_pointcloud [default: true]
      • use_pointcloud_container [default: true]
  • launch/traffic_light_recognition/traffic_light_detector_fine.launch.xml
    • Fine-detection chain for one traffic light camera: map based detector proposes rough ROIs, the fine detector refines them, classifiers label them.
      • perception_config_pkg [default: autoware_perception_config]
      • camera_namespace
      • enable_image_decompressor [default: true]
      • input/vector_map [default: /map/vector_map]
      • input/route [default: /planning/mission_planning/route]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • map_based_detector_param_path [default: $(find-pkg-share autoware_traffic_light_map_based_detector)/config/traffic_light_map_based_detector.param.yaml]
      • fine_detector_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_fine_detector/traffic_light_fine_detector.param.yaml]
      • car_classifier_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_classifier/car_traffic_light_classifier.param.yaml]
      • pedestrian_classifier_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_classifier/pedestrian_traffic_light_classifier.param.yaml]
      • lamp_recognizer_ml_param_path [default: $(var data_path)/traffic_light_classifier/lamp_recognizer_ml.param.yaml]
      • roi_visualizer_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_visualization/traffic_light_roi_visualizer.param.yaml]
      • fine_detector_model_path [default: $(var data_path)/traffic_light_fine_detector/tlr_car_ped_yolox_s_batch_6.onnx]
      • fine_detector_label_path [default: $(var data_path)/traffic_light_fine_detector/tlr_labels.txt]
      • car_classifier_model_path [default: $(var data_path)/traffic_light_classifier/traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • car_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels.txt]
      • pedestrian_classifier_model_path [default: $(var data_path)/traffic_light_classifier/ped_traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • pedestrian_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels_ped.txt]
      • car_classifier_type [default: 1]
      • pedestrian_classifier_type [default: 1]
  • launch/traffic_light_recognition/traffic_light_detector_rough.launch.xml
    • Rough-detection chain for one traffic light camera: map based detector ROIs feed the classifiers directly, without a high accuracy detector.
      • perception_config_pkg [default: autoware_perception_config]
      • camera_namespace
      • enable_image_decompressor [default: true]
      • input/vector_map [default: /map/vector_map]
      • input/route [default: /planning/mission_planning/route]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • map_based_detector_param_path [default: $(find-pkg-share autoware_traffic_light_map_based_detector)/config/traffic_light_map_based_detector.param.yaml]
      • car_classifier_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_classifier/car_traffic_light_classifier.param.yaml]
      • pedestrian_classifier_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_classifier/pedestrian_traffic_light_classifier.param.yaml]
      • lamp_recognizer_ml_param_path [default: $(var data_path)/traffic_light_classifier/lamp_recognizer_ml.param.yaml]
      • roi_visualizer_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_visualization/traffic_light_roi_visualizer.param.yaml]
      • car_classifier_model_path [default: $(var data_path)/traffic_light_classifier/traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • car_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels.txt]
      • pedestrian_classifier_model_path [default: $(var data_path)/traffic_light_classifier/ped_traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • pedestrian_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels_ped.txt]
      • car_classifier_type [default: 1]
      • pedestrian_classifier_type [default: 1]
  • launch/traffic_light_recognition/traffic_light_detector_whole_image.launch.xml
    • Whole-image detection chain for one traffic light camera: yolox detects over the full image, the selector matches detections to map based ROIs, classifiers label them.
      • perception_config_pkg [default: autoware_perception_config]
      • camera_namespace
      • enable_image_decompressor [default: true]
      • input/vector_map [default: /map/vector_map]
      • input/route [default: /planning/mission_planning/route]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • map_based_detector_param_path [default: $(find-pkg-share autoware_traffic_light_map_based_detector)/config/traffic_light_map_based_detector.param.yaml]
      • whole_image_detector_param_path [default: $(var perception_config_dir)/traffic_light_recognition/tensorrt_yolox/yolox_traffic_light_detector.param.yaml]
      • car_classifier_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_classifier/car_traffic_light_classifier.param.yaml]
      • pedestrian_classifier_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_classifier/pedestrian_traffic_light_classifier.param.yaml]
      • lamp_recognizer_ml_param_path [default: $(var data_path)/traffic_light_classifier/lamp_recognizer_ml.param.yaml]
      • roi_visualizer_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_visualization/traffic_light_roi_visualizer.param.yaml]
      • whole_image_detector_model_path [default: $(var data_path)/tensorrt_yolox/yolox_s_car_ped_tl_detector_960_960_batch_1.onnx]
      • whole_image_detector_label_path [default: $(var data_path)/tensorrt_yolox/car_ped_tl_detector_labels.txt]
      • car_classifier_model_path [default: $(var data_path)/traffic_light_classifier/traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • car_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels.txt]
      • pedestrian_classifier_model_path [default: $(var data_path)/traffic_light_classifier/ped_traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • pedestrian_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels_ped.txt]
      • car_classifier_type [default: 1]
      • pedestrian_classifier_type [default: 1]
  • launch/traffic_light_recognition/traffic_light_fusion.launch.xml
    • Post-process shared by all traffic light cameras: multi camera fusion, V2X arbitration, pedestrian signal estimation and map visualization.
      • perception_config_pkg [default: autoware_perception_config]
      • camera_namespaces [default: [camera6, camera7]]
      • input/vector_map [default: /map/vector_map]
      • external/traffic_signals [default: /perception/traffic_light_recognition/external/traffic_signals]
      • internal/traffic_signals [default: /perception/traffic_light_recognition/internal/traffic_signals]
      • judged/traffic_signals [default: /perception/traffic_light_recognition/judged/traffic_signals]
      • output/traffic_signals [default: /perception/traffic_light_recognition/traffic_signals]
      • multi_camera_fusion_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_multi_camera_fusion/traffic_light_multi_camera_fusion.param.yaml]
      • arbiter_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_arbiter/traffic_light_arbiter.param.yaml]
      • crosswalk_estimator_param_path [default: $(var perception_config_dir)/traffic_light_recognition/crosswalk_traffic_light_estimator/crosswalk_traffic_light_estimator.param.yaml]
  • launch/traffic_light_recognition/traffic_light_occlusion_predictor.launch.xml
    • Occlusion check for one traffic light camera: merges the car and pedestrian signals into the per-camera output, masking signals hidden from the camera.
      • perception_config_pkg [default: autoware_perception_config]
      • camera_namespace
      • input/cloud [default: /sensing/lidar/top/pointcloud_raw_ex]
      • input/vector_map [default: /map/vector_map]
      • occlusion_predictor_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_occlusion_predictor/traffic_light_occlusion_predictor.param.yaml]
  • launch/traffic_light_recognition/traffic_light_recognition_fine.launch.xml
    • Traffic light recognition, fine detection mode: per-camera fine detection with occlusion check, then shared fusion. Two cameras as the representative set; add or remove per-camera blocks to match the vehicle.
      • perception_config_pkg [default: autoware_perception_config]
      • input/vector_map [default: /map/vector_map]
      • input/route [default: /planning/mission_planning/route]
      • input/pointcloud [default: /sensing/lidar/top/pointcloud_raw_ex]
      • external/traffic_signals [default: /perception/traffic_light_recognition/external/traffic_signals]
      • output/traffic_signals [default: /perception/traffic_light_recognition/traffic_signals]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • fine_detector_model_path [default: $(var data_path)/traffic_light_fine_detector/tlr_car_ped_yolox_s_batch_6.onnx]
      • fine_detector_label_path [default: $(var data_path)/traffic_light_fine_detector/tlr_labels.txt]
      • lamp_recognizer_ml_param_path [default: $(var data_path)/traffic_light_classifier/lamp_recognizer_ml.param.yaml]
      • car_classifier_model_path [default: $(var data_path)/traffic_light_classifier/traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • car_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels.txt]
      • pedestrian_classifier_model_path [default: $(var data_path)/traffic_light_classifier/ped_traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • pedestrian_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels_ped.txt]
      • car_classifier_type [default: 1]
      • pedestrian_classifier_type [default: 1]
      • base_map_based_detector_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_map_based_detector/base_traffic_light_map_based_detector.param.yaml]
      • extra_map_based_detector_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_map_based_detector/extra_traffic_light_map_based_detector.param.yaml]
  • launch/traffic_light_recognition/traffic_light_recognition_merge.launch.xml
    • Traffic light recognition, merge mode: detection and classification run on other ECUs; this side runs the per-camera occlusion check and the shared fusion. Two cameras as the representative set; add or remove per-camera blocks to match the vehicle.
      • input/vector_map [default: /map/vector_map]
      • input/pointcloud [default: /sensing/lidar/top/pointcloud_raw_ex]
      • external/traffic_signals [default: /perception/traffic_light_recognition/external/traffic_signals]
      • output/traffic_signals [default: /perception/traffic_light_recognition/traffic_signals]
  • launch/traffic_light_recognition/traffic_light_recognition_rough.launch.xml
    • Traffic light recognition, rough detection mode: per-camera map based detection with occlusion check, then shared fusion. One camera as the representative set; add or remove per-camera blocks to match the vehicle.
      • perception_config_pkg [default: autoware_perception_config]
      • input/vector_map [default: /map/vector_map]
      • input/route [default: /planning/mission_planning/route]
      • input/pointcloud [default: /sensing/lidar/top/pointcloud_raw_ex]
      • external/traffic_signals [default: /perception/traffic_light_recognition/external/traffic_signals]
      • output/traffic_signals [default: /perception/traffic_light_recognition/traffic_signals]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • lamp_recognizer_ml_param_path [default: $(var data_path)/traffic_light_classifier/lamp_recognizer_ml.param.yaml]
      • car_classifier_model_path [default: $(var data_path)/traffic_light_classifier/traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • car_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels.txt]
      • pedestrian_classifier_model_path [default: $(var data_path)/traffic_light_classifier/ped_traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • pedestrian_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels_ped.txt]
      • car_classifier_type [default: 1]
      • pedestrian_classifier_type [default: 1]
      • base_map_based_detector_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_map_based_detector/base_traffic_light_map_based_detector.param.yaml]
  • launch/traffic_light_recognition/traffic_light_recognition_whole_image.launch.xml
    • Traffic light recognition, whole-image detection mode: per-camera yolox detection, then shared fusion. Two cameras as the representative set; add or remove per-camera blocks to match the vehicle.
      • perception_config_pkg [default: autoware_perception_config]
      • input/vector_map [default: /map/vector_map]
      • input/route [default: /planning/mission_planning/route]
      • external/traffic_signals [default: /perception/traffic_light_recognition/external/traffic_signals]
      • output/traffic_signals [default: /perception/traffic_light_recognition/traffic_signals]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • whole_image_detector_model_path [default: $(var data_path)/tensorrt_yolox/yolox_s_car_ped_tl_detector_960_960_batch_1.onnx]
      • whole_image_detector_label_path [default: $(var data_path)/tensorrt_yolox/car_ped_tl_detector_labels.txt]
      • lamp_recognizer_ml_param_path [default: $(var data_path)/traffic_light_classifier/lamp_recognizer_ml.param.yaml]
      • car_classifier_model_path [default: $(var data_path)/traffic_light_classifier/traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • car_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels.txt]
      • pedestrian_classifier_model_path [default: $(var data_path)/traffic_light_classifier/ped_traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • pedestrian_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels_ped.txt]
      • car_classifier_type [default: 1]
      • pedestrian_classifier_type [default: 1]
      • base_map_based_detector_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_map_based_detector/base_traffic_light_map_based_detector.param.yaml]
      • extra_map_based_detector_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_map_based_detector/extra_traffic_light_map_based_detector.param.yaml]

Messages

No message files found.

Services

No service files found

Plugins

No plugins found.

Recent questions tagged autoware_perception_launch at Robotics Stack Exchange

No version for distro lyrical showing github. Known supported distros are highlighted in the buttons above.
Package symbol

autoware_perception_launch package from autoware_launch repo

autoware_launch autoware_sample_designs autoware_control_config autoware_control_launch autoware_localization_config autoware_localization_launch autoware_map_config autoware_map_launch autoware_perception_config autoware_perception_launch autoware_planning_config autoware_planning_launch autoware_sensing_launch autoware_sensor_driver_launch awsim_labs_sensor_kit_description awsim_labs_sensor_kit_launch awsim_sensor_kit_description awsim_sensor_kit_launch carla_sensor_kit_description carla_sensor_kit_launch sample_sensor_kit_description sample_sensor_kit_launch single_lidar_sensor_kit_description single_lidar_sensor_kit_launch tier4_autoware_api_launch tier4_simulator_launch tier4_system_launch tier4_vehicle_launch awsim_labs_vehicle_description awsim_labs_vehicle_launch sample_vehicle_description sample_vehicle_launch

ROS Distro
github

API Docs
Browse Code

Package Summary

Version 0.53.0
License Apache License 2.0
Build type AMENT_CMAKE
Use RECOMMENDED

Repository Summary

Description
Checkout URI https://github.com/autowarefoundation/autoware_launch.git
VCS Type git
VCS Version main
Last Updated 2026-10-26
Dev Status UNKNOWN
Released UNRELEASED
Contributing Help Wanted (-)
Good First Issues (-)
Pull Requests to Review (-)

Package Description

The autoware_perception_launch package

Maintainers

  • Yukihiro Saito
  • Dai Nguyen
  • Taekjin Lee
  • Yoshi Ri
  • Masato Saeki

Authors

No additional authors.
autoware_universe_launch/autoware_perception_launch/README.md

autoware_perception_launch

Shared perception launch library. The system component launchers live in the product layer (autoware_launch/launch/components/component_perception.launch.xml for the lidar pipeline and component_traffic_light.launch.xml for traffic light recognition); this package hosts the perception feature launchers they combine:

launch/
├── common/                        # shared preprocess feature launchers (pointcloud downsample)
├── object_recognition/
│   ├── object_recognition_<mode>.launch.xml  # detection + tracking + prediction per sensor mode
│   ├── detection/
│   │   ├── detector/              # detectors
│   │   ├── filter/                # filters
│   │   └── merger/                # detection-side serial merger (used by the camera_lidar_radar_serial launcher)
│   └── prediction/
├── obstacle_segmentation/
├── occupancy_grid_map/
└── traffic_light_recognition/
design/module/                     # Autoware System Designer modules; same subtree as launch/
├── object_recognition/
├── obstacle_segmentation/
├── occupancy_grid_map/
└── traffic_light_recognition/

Launch files

  • object_recognition/object_recognition_<mode>.launch.xml — one feature launcher per sensor mode (camera, lidar, lidar_ptv3, lidar_radar, camera_lidar_radar, camera_lidar_radar_serial); hosts the detection stage inline together with the tracker and prediction, and pushes their module namespaces. Camera and radar input topics default to their absolute sensing paths (/sensing/camera/cameraN/..., /sensing/radar/detected_objects), so callers override only what differs. Most modes wire the multi-channel tracker to the detector outputs; camera_lidar_radar_serial instead merges the detectors in the detection stage (optionally regenerating detections through detection-by-tracker), tracks the single merged detected_objects channel, and joins radar tracked objects at the decorative tracker merger. camera runs camera-only detectors (BEV and camera VRU).
  • The clustering backend is file selection, not a flag: lidar runs rule-based euclidean clustering off the obstacle segmentation cloud, while lidar_ptv3 and lidar_radar cluster the PTv3 semantic-segmentation cloud and drop the pointcloud map filter and the object validator that only the rule-based pipeline uses. Both feed their cluster objects to the tracker on the lidar_clustering channel.
  • object_recognition/detection/{detector,filter}/ — single-purpose launchers wired together by the object-recognition launchers. merger/camera_lidar_merger.launch.xml chains the detection-side object mergers for the camera_lidar_radar_serial launcher.
  • obstacle_segmentation/ground_segmentation_<backend>.launch.xml — ground removal backends (rule_based, rule_based_cuda, ml); backend selection is file selection. ground_segmentation_{additional_lidars,ransac}.launch.xml fold extra lidars into the obstacle stream for kits that carry them.
  • occupancy_grid_map/, common/, traffic_light_recognition/ — feature launchers included directly by the system components.

Namespace and container addressing

Callers own the namespaces: the system component pushes perception and object_recognition; object_recognition_<mode>.launch.xml pushes detection/tracking/prediction. The object-recognition launchers address their internal topics through the fixed /perception/object_recognition namespace. Composable nodes load into the shared pointcloud container, whose name is exposed as pointcloud_container_name and aliased to the node/pointcloud_container argument of the detector launchers.

Config

Parameter files are resolved from a config/ tree addressed through the perception_config_pkg argument, which defaults to autoware_perception_config. A product can relocate the whole tree by setting perception_config_pkg once at its entry point — the launch configuration propagates through every include — but it must then ship the same config/ layout. Products that differ in only a few files instead override the corresponding *_param_path arguments, which every feature launcher declares with a default:

<include file="$(find-pkg-share autoware_perception_launch)/launch/object_recognition/object_recognition_lidar.launch.xml">
  <arg name="perception_config_pkg" value="my_perception_config"/>
  <arg name="FOO_NODE_param_path" value="..."/>
</include>

component_perception.launch.xml, component_traffic_light.launch.xml and planning_simulator/perception_simulation.launch.xml each declare perception_config_pkg, so a product switches the tree at the component it launches.

The launch/ tree is the interface downstream distributions adopt as-is; only package.xml, CMakeLists.txt, README.md, and changelogs stay package-local. Keep changes to this tree folder-merge compatible.

Package Dependencies

Please see <exec_depend> in package.xml.

CHANGELOG

Changelog for package autoware_perception_launch

0.53.0 (2026-10-13)

  • Merge remote-tracking branch 'origin/main' into tmp/bot/bump_version_base

  • feat(autoware_launch): run crop_box_filter standalone when agnocast is enabled (#1990)

    * feat(autoware_launch): run crop_box_filter standalone when agnocast is enabled autoware_universe#13384 moves CropBoxFilterComponent to agnocast_wrapper::Node on an agnocast-only executor, which the component containers do not set up. Start each of the eight crop_box_filter nodes as a standalone process when ENABLE_AGNOCAST=1 and keep loading them into their container otherwise. The node names, namespaces, remaps and parameters are the same in both branches.

    * refactor(autoware_launch): launch crop_box_filter through the autoware_node action ---------

  • feat(autoware_perception_launch): run scan_ground_filter standalone when agnocast is enabled (#1991)

    * feat(autoware_perception_launch): run scan_ground_filter standalone when agnocast is enabled autoware_universe#13385 moves ScanGroundFilterComponent to agnocast_wrapper::Node on an agnocast-only executor, which the component containers do not set up. Start each of the three scan_ground_filter nodes as a standalone process when ENABLE_AGNOCAST=1 and keep loading them into their container otherwise. The node names, namespaces, remaps and parameters are the same in both branches.

    * refactor(autoware_perception_launch): launch scan_ground_filter through the autoware_node action ---------

  • feat(autoware_launch): run voxel_grid_downsample_filter standalone when agnocast is enabled (#1992)

    • feat(autoware_launch): run voxel_grid_downsample_filter standalone when agnocast is enabled
    • style(pre-commit): autofix

    * refactor(autoware_launch): launch voxel_grid_downsample_filter through the autoware_node action ---------Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>

  • feat(autoware_sample_designs): compose the AD API, system and vehicle components from design modules (#1976)

    * fix(control, planning): connect the operation mode state and velocity limit inputs Wires collision_detector's operation_mode_state and remaining_distance_time_calculator's planning_velocity through the module designs; their node designs now declare the fixed topic names as remap targets.

    * fix(planning): connect the internal velocity limit path to the selector Forwards motion_velocity_planner's velocity_limit and clear_velocity_limit up to ScenarioPlanning and feeds them into external_velocity_limit_selector, whose internal inputs were unconnected.

    * fix(control, planning): connect the remaining module inputs to their sources Wires the operation mode transition manager's gate and selector inputs, the stop mode operator's vehicle status, the manual lane change handler's route availability and preferred primitive service, the parking costmap, and the route marker. Exposes the vehicle control mode service, the preferred lane service and the system hazard lights command as component interfaces.

    • fix(planning): follow the hazard lights selector input rename
    • fix(planning): add steering_status subscriber to various planning modules
    • feat(vehicle): add vehicle wrapper and launch file for vehicle interface
    • style(pre-commit): autofix
    • fix(planning): add planning factors publishers and connections for various modules
    • refactor(autoware_sample_designs): compose the AD API from design modules
    • Refactor ADAPI module structure and enhance planning factors
    • Updated AdapiRvizHelpers.module.yaml to include helper prefixes for instances and connections.
    • Modified AutowareSample.system.yaml to reorganize ADAPI components, introducing default_adapi for better structure.
    • Expanded planning factors in Planning.module.yaml, LaneDriving.module.yaml, ScenarioPlanning.module.yaml, and associated modules to include additional factors for improved behavior and motion planning.
    • Enhanced connections in behavior planning and motion planning modules to accommodate new planning factors.
    • Ensured consistency across scenario planning modules by aligning

File truncated at 100 lines see the full file

Package Dependencies

Deps Name
ament_cmake_auto
autoware_cmake
autoware_agnocast_wrapper
autoware_bevfusion
autoware_bytetrack
autoware_compare_map_segmentation
autoware_crosswalk_traffic_light_estimator
autoware_cuda_pointcloud_preprocessor
autoware_detected_object_feature_remover
autoware_detected_object_validation
autoware_detection_by_tracker
autoware_elevation_map_loader
autoware_euclidean_cluster
autoware_ground_segmentation
autoware_ground_segmentation_cuda
autoware_image_object_locator
autoware_image_projection_based_fusion
autoware_image_transport_decompressor
autoware_lidar_apollo_instance_segmentation
autoware_lidar_centerpoint
autoware_lidar_transfusion
autoware_map_based_prediction
autoware_multi_object_tracker
autoware_object_merger
autoware_object_sorter
autoware_occupancy_grid_map_outlier_filter
autoware_perception_config
autoware_pointcloud_preprocessor
autoware_probabilistic_occupancy_grid_map
autoware_ptv3
autoware_raindrop_cluster_filter
autoware_shape_estimation
autoware_simpl_prediction
autoware_simple_object_merger
autoware_tensorrt_bevdet
autoware_tensorrt_yolox
autoware_tracking_object_merger
autoware_traffic_light_arbiter
autoware_traffic_light_category_merger
autoware_traffic_light_classifier
autoware_traffic_light_fine_detector
autoware_traffic_light_map_based_detector
autoware_traffic_light_multi_camera_fusion
autoware_traffic_light_occlusion_predictor
autoware_traffic_light_selector
autoware_traffic_light_visualization
rclcpp_components
topic_tools
ament_lint_auto
autoware_lint_common

System Dependencies

No direct system dependencies.

Dependant Packages

Name Deps
autoware_launch
reaction_analyzer

Launch files

  • launch/common/pointcloud_downsample.launch.xml
      • pointcloud_container_name
      • cuda_pointcloud_preprocessing [default: false]
      • input/pointcloud
      • output/pointcloud
      • common_downsample_voxel_size_x [default: 0.05]
      • common_downsample_voxel_size_y [default: 0.05]
      • common_downsample_voxel_size_z [default: 0.05]
  • launch/object_recognition/detection/detector/camera_2d_detector.launch.xml
      • image_raw0 [default: /sensing/camera/camera0/image_raw]
      • image_raw1 [default: /sensing/camera/camera1/image_raw]
      • image_raw2 [default: /sensing/camera/camera2/image_raw]
      • image_raw3 [default: /sensing/camera/camera3/image_raw]
      • image_raw4 [default: /sensing/camera/camera4/image_raw]
      • image_raw5 [default: /sensing/camera/camera5/image_raw]
      • image_raw6 [default: /sensing/camera/camera6/image_raw]
      • image_raw7 [default: /sensing/camera/camera7/image_raw]
      • image_raw8 [default: /sensing/camera/camera8/image_raw]
      • image_raw9 [default: /sensing/camera/camera9/image_raw]
      • image_number [default: 1]
      • camera_index [default: 0]
      • use_bytetrack [default: true]
      • enable_visualizer [default: false]
      • camera_2d_detector/model_path
      • camera_2d_detector/label_path
      • camera_2d_detector/color_map_path
      • camera_2d_detector/roi_remap_path
      • camera_2d_detector/roi_to_semseg_remap_path
      • tensorrt_yolox_ns [default: ]
  • launch/object_recognition/detection/detector/camera_bev_detector.launch.xml
      • input/camera0/image
      • input/camera0/info
      • input/camera1/image
      • input/camera1/info
      • input/camera2/image
      • input/camera2/info
      • input/camera3/image
      • input/camera3/info
      • input/camera4/image
      • input/camera4/info
      • input/camera5/image
      • input/camera5/info
      • output/objects
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • bevdet_model_name [default: bevdet_one_lt_d]
      • bevdet_model_path [default: $(var data_path)/tensorrt_bevdet]
  • launch/object_recognition/detection/detector/camera_lidar_detector.launch.xml
      • ns
      • lidar_detection_model_type
      • lidar_detection_model_name
      • use_low_intensity_cluster_filter
      • use_image_segmentation_based_filter
      • number_of_cameras
      • input/camera0/image
      • input/camera0/info
      • input/camera0/rois
      • input/camera1/image
      • input/camera1/info
      • input/camera1/rois
      • input/camera2/image
      • input/camera2/info
      • input/camera2/rois
      • input/camera3/image
      • input/camera3/info
      • input/camera3/rois
      • input/camera4/image
      • input/camera4/info
      • input/camera4/rois
      • input/camera5/image
      • input/camera5/info
      • input/camera5/rois
      • input/camera6/image
      • input/camera6/info
      • input/camera6/rois
      • input/camera7/image
      • input/camera7/info
      • input/camera7/rois
      • input/camera8/image
      • input/camera8/info
      • input/camera8/rois
      • input/concatenation_info
      • segmentation_pointcloud_fusion_camera_ids
      • image_topic_name
      • sync_param_path
      • voxel_grid_based_euclidean_param_path
      • lidar_model_param_path
      • pointpainting_model_path [default: $(var data_path)/image_projection_based_fusion]
      • pointpainting_fusion_common_param_path
      • roi_cluster_fusion_param_path
      • node/pointcloud_container
      • input/pointcloud
      • input/pointcloud_map/pointcloud
      • input/obstacle_segmentation/pointcloud
      • output/ml_detector/objects
      • output/rule_detector/objects
      • camera_2d_detector/model_path
      • camera_2d_detector/label_path
      • camera_2d_detector/color_map_path
      • camera_2d_detector/roi_remap_path
      • camera_2d_detector/roi_to_semseg_remap_path
      • enable_2d_detection [default: false]
  • launch/object_recognition/detection/detector/camera_lidar_irregular_object_detector.launch.xml
      • pipeline_ns
      • node/pointcloud_container
      • output/objects [default: objects]
      • input/concatenation_info
      • input/pointcloud
      • irregular_object_crop_box_filter_node_param_path
      • irregular_object_scan_ground_filter_node_param_path
      • irregular_object_roi_pointcloud_fusion_node_param_path
      • irregular_object_fusion_camera_param_path
      • sync_param_path
  • launch/object_recognition/detection/detector/camera_vru_detector.launch.xml
      • input/camera0/info [default: /sensing/camera/camera0/camera_info]
      • input/camera0/rois [default: /perception/object_recognition/detection/rois0]
      • input/camera1/info [default: /sensing/camera/camera1/camera_info]
      • input/camera1/rois [default: /perception/object_recognition/detection/rois1]
      • input/camera2/info [default: /sensing/camera/camera2/camera_info]
      • input/camera2/rois [default: /perception/object_recognition/detection/rois2]
      • input/camera3/info [default: /sensing/camera/camera3/camera_info]
      • input/camera3/rois [default: /perception/object_recognition/detection/rois3]
      • input/camera4/info [default: /sensing/camera/camera4/camera_info]
      • input/camera4/rois [default: /perception/object_recognition/detection/rois4]
      • input/camera5/info [default: /sensing/camera/camera5/camera_info]
      • input/camera5/rois [default: /perception/object_recognition/detection/rois5]
      • input/camera6/info [default: /sensing/camera/camera6/camera_info]
      • input/camera6/rois [default: /perception/object_recognition/detection/rois6]
      • input/camera7/info [default: /sensing/camera/camera7/camera_info]
      • input/camera7/rois [default: /perception/object_recognition/detection/rois7]
      • output/objects [default: /perception/object_recognition/detection/camera_vru/objects]
      • bbox_object_locator_param_path
      • rois_ids [default: [0, 1]]
  • launch/object_recognition/detection/detector/irregular_object_detector.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • target_container
      • base_frame [default: base_link]
      • use_intra_process [default: true]
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • input/concatenation_info [default: /sensing/lidar/concatenated/pointcloud_info]
      • output/clusters
      • crop_box_filter_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/crop_box_filter.param.yaml]
      • scan_ground_filter_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/scan_ground_filter.param.yaml]
      • roi_pointcloud_fusion_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/roi_pointcloud_fusion.param.yaml]
      • fusion_camera_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/fusion_cameras.param.yaml]
      • sync_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/fusion_common.param.yaml]
  • launch/object_recognition/detection/detector/lidar_dnn_detector.launch.xml
      • lidar_detection_model_type
      • lidar_detection_model_name
      • lidar_model_param_path
      • bevfusion_model_path [default: $(var data_path)/bevfusion]
      • centerpoint_model_path [default: $(var data_path)/lidar_centerpoint]
      • centerpoint_common_param_path [default: $(find-pkg-share autoware_lidar_centerpoint)/config/centerpoint_common.param.yaml]
      • transfusion_model_path [default: $(var data_path)/lidar_transfusion]
      • use_short_range_detection [default: false]
      • lidar_short_range_detection_model_type
      • node/pointcloud_container
      • input/pointcloud
      • output/objects
      • output/short_range_objects
  • launch/object_recognition/detection/detector/lidar_rule_detector.launch.xml
      • ns
      • node/pointcloud_container
      • input/pointcloud_map/pointcloud
      • input/obstacle_segmentation/pointcloud
      • output/cluster_objects
      • output/objects
      • voxel_grid_based_euclidean_param_path
      • use_low_height_cropbox [default: false]
  • launch/object_recognition/detection/detector/lidar_semantic_segmentation_detector.launch.xml
      • euclidean_clustering/param_path
      • input/pointcloud
      • output/objects
  • launch/object_recognition/detection/detector/tracker_based_detector.launch.xml
      • detection_by_tracker_param_path
      • input/clusters
      • input/tracked_objects
      • output/objects
  • launch/object_recognition/detection/filter/object_filter.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • objects_filter_method [default: lanelet_filter]
      • object_recognition_detection_object_lanelet_filter_param_path [default: $(var perception_config_dir)/object_recognition/detection/object_filter/object_lanelet_filter.param.yaml]
      • object_recognition_detection_object_position_filter_param_path [default: $(var perception_config_dir)/object_recognition/detection/object_filter/object_position_filter.param.yaml]
      • input/objects
      • output/objects
  • launch/object_recognition/detection/filter/object_validator.launch.xml
      • objects_validation_method
      • object_recognition_detection_obstacle_pointcloud_based_validator_param_path
      • input/obstacle_pointcloud
      • input/objects
      • output/objects
  • launch/object_recognition/detection/filter/pointcloud_downsample_filter.launch.xml
      • target_container
      • use_intra_process [default: true]
      • input/pointcloud
      • output/pointcloud
      • downsample_voxel_size [default: 0.1]
  • launch/object_recognition/detection/filter/pointcloud_map_filter.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • target_container
      • use_intra_process [default: true]
      • input/pointcloud
      • output/pointcloud
      • output/downsampled/pointcloud [default: /perception/obstacle_segmentation/pointcloud_map_filtered/downsampled/pointcloud]
      • downsample_voxel_size [default: 0.1]
      • voxel_based_compare_map_filter_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/pointcloud_filter/voxel_based_compare_map_filter.param.yaml]
  • launch/object_recognition/detection/filter/radar_filter.launch.xml
      • object_sorter_param_path
      • radar_lanelet_filtering_range_param_path
      • input/radar
      • output/objects
  • launch/object_recognition/detection/merger/camera_lidar_merger.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • object_recognition_detection_object_merger_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/detection/object_merger/data_association_matrix.param.yaml]
      • object_recognition_detection_object_merger_distance_threshold_list_path [default: $(var perception_config_dir)/object_recognition/detection/object_merger/overlapped_judge.param.yaml]
      • ml_camera_lidar_object_association_merger_param_path [default: $(var perception_config_dir)/object_recognition/detection/object_merger/ml_camera_lidar_object_association_merger.param.yaml]
      • roi_detected_object_fusion_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/roi_detected_object_fusion.param.yaml]
      • sync_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/fusion_common.param.yaml]
      • lidar_detection_model_type
      • use_detection_by_tracker
      • use_irregular_object_detector
      • use_roi_detected_object_validator
      • euclidean_clustering/shape_policy
      • use_object_filter
      • objects_filter_method
      • number_of_cameras
      • input/camera0/image
      • input/camera0/info
      • input/camera0/rois
      • input/camera1/image
      • input/camera1/info
      • input/camera1/rois
      • input/camera2/image
      • input/camera2/info
      • input/camera2/rois
      • input/camera3/image
      • input/camera3/info
      • input/camera3/rois
      • input/camera4/image
      • input/camera4/info
      • input/camera4/rois
      • input/camera5/image
      • input/camera5/info
      • input/camera5/rois
      • input/camera6/image
      • input/camera6/info
      • input/camera6/rois
      • input/camera7/image
      • input/camera7/info
      • input/camera7/rois
      • input/camera8/image
      • input/camera8/info
      • input/camera8/rois
      • input/concatenation_info [default: /sensing/lidar/concatenated/pointcloud_info]
      • input/lidar_ml/objects
      • input/lidar_cluster/objects
      • input/irregular_object/objects [default: ]
      • input/detection_by_tracker/objects
      • output/objects [default: objects]
      • alpha_merger_priority_mode [default: 0]
  • launch/object_recognition/object_recognition_camera.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • use_camera_vru_detector [default: false]
      • camera_vru_detector_rois_ids [default: [0]]
      • use_camera_bev_detector [default: true]
      • use_vector_map [default: true]
      • tracker_publish_merged_objects [default: false]
      • input/camera0/image [default: /sensing/camera/camera0/image_rect_color]
      • input/camera0/info [default: /sensing/camera/camera0/camera_info]
      • input/camera0/rois [default: /perception/object_recognition/detection/rois0]
      • input/camera1/image [default: /sensing/camera/camera1/image_rect_color]
      • input/camera1/info [default: /sensing/camera/camera1/camera_info]
      • input/camera1/rois [default: /perception/object_recognition/detection/rois1]
      • input/camera2/image [default: /sensing/camera/camera2/image_rect_color]
      • input/camera2/info [default: /sensing/camera/camera2/camera_info]
      • input/camera2/rois [default: /perception/object_recognition/detection/rois2]
      • input/camera3/image [default: /sensing/camera/camera3/image_rect_color]
      • input/camera3/info [default: /sensing/camera/camera3/camera_info]
      • input/camera3/rois [default: /perception/object_recognition/detection/rois3]
      • input/camera4/image [default: /sensing/camera/camera4/image_rect_color]
      • input/camera4/info [default: /sensing/camera/camera4/camera_info]
      • input/camera4/rois [default: /perception/object_recognition/detection/rois4]
      • input/camera5/image [default: /sensing/camera/camera5/image_rect_color]
      • input/camera5/info [default: /sensing/camera/camera5/camera_info]
      • input/camera5/rois [default: /perception/object_recognition/detection/rois5]
      • input/camera6/image [default: /sensing/camera/camera6/image_rect_color]
      • input/camera6/info [default: /sensing/camera/camera6/camera_info]
      • input/camera6/rois [default: /perception/object_recognition/detection/rois6]
      • input/camera7/image [default: /sensing/camera/camera7/image_rect_color]
      • input/camera7/info [default: /sensing/camera/camera7/camera_info]
      • input/camera7/rois [default: /perception/object_recognition/detection/rois7]
      • output/tracker_merged_objects [default: /perception/object_recognition/detection/objects]
      • output/tracked_objects [default: /perception/object_recognition/tracking/objects]
      • output/predicted_objects [default: /perception/object_recognition/objects]
      • near_range_camera_vru_param_path [default: $(var perception_config_dir)/object_recognition/detection/camera_vru_detection/near_range_camera_vru_detector.param.yaml]
      • object_recognition_tracking_multi_object_tracker_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/data_association_matrix.param.yaml]
      • object_recognition_tracking_multi_object_tracker_input_channels_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/input_channels.param.yaml]
      • object_recognition_tracking_multi_object_tracker_node_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/multi_object_tracker_node.param.yaml]
  • launch/object_recognition/object_recognition_camera_lidar_radar.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • lidar_detection_model [default: centerpoint]
      • use_short_range_detection [default: false]
      • lidar_short_range_detection_model_type [default: centerpoint_short_range]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • bevfusion_model_path [default: $(var data_path)/bevfusion]
      • pointpainting_model_path [default: $(var data_path)/image_projection_based_fusion]
      • camera_2d_detector/model_path [default: $(var data_path)/tensorrt_yolox/yolox-sPlus-opt-pseudoV2-T4-960x960-T4-seg16cls.onnx]
      • camera_2d_detector/label_path [default: $(var data_path)/tensorrt_yolox/label.txt]
      • camera_2d_detector/color_map_path [default: $(var data_path)/tensorrt_yolox/semseg_color_map.csv]
      • camera_2d_detector/roi_remap_path [default: $(var perception_config_dir)/traffic_light_recognition/tensorrt_yolox/roi_label_remap.csv]
      • camera_2d_detector/roi_to_semseg_remap_path [default: $(var perception_config_dir)/traffic_light_recognition/tensorrt_yolox/roi_to_semseg_label_remap.csv]
      • use_semantic_segmentation_ptv3 [default: false]
      • use_object_filter [default: true]
      • use_pointcloud_map [default: true]
      • use_validator [default: false]
      • objects_validation_method [default: obstacle_pointcloud]
      • use_low_height_cropbox [default: false]
      • use_low_intensity_cluster_filter [default: true]
      • use_image_segmentation_based_filter [default: false]
      • use_irregular_object_detector [default: true]
      • use_camera_vru_detector [default: false]
      • camera_vru_detector_rois_ids [default: [0]]
      • use_camera_only_detection [default: false]
      • use_vector_map [default: true]
      • tracker_publish_merged_objects [default: false]
      • number_of_cameras [default: 6]
      • pointcloud_container_name [default: pointcloud_container]
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • input/semantic_segmentation/pointcloud [default: /perception/segmented/pointcloud]
      • input/obstacle_segmentation/pointcloud [default: /perception/obstacle_segmentation/pointcloud]
      • input/camera0/image [default: /sensing/camera/camera0/image_rect_color]
      • input/camera0/info [default: /sensing/camera/camera0/camera_info]
      • input/camera0/rois [default: /perception/object_recognition/detection/rois0]
      • input/camera1/image [default: /sensing/camera/camera1/image_rect_color]
      • input/camera1/info [default: /sensing/camera/camera1/camera_info]
      • input/camera1/rois [default: /perception/object_recognition/detection/rois1]
      • input/camera2/image [default: /sensing/camera/camera2/image_rect_color]
      • input/camera2/info [default: /sensing/camera/camera2/camera_info]
      • input/camera2/rois [default: /perception/object_recognition/detection/rois2]
      • input/camera3/image [default: /sensing/camera/camera3/image_rect_color]
      • input/camera3/info [default: /sensing/camera/camera3/camera_info]
      • input/camera3/rois [default: /perception/object_recognition/detection/rois3]
      • input/camera4/image [default: /sensing/camera/camera4/image_rect_color]
      • input/camera4/info [default: /sensing/camera/camera4/camera_info]
      • input/camera4/rois [default: /perception/object_recognition/detection/rois4]
      • input/camera5/image [default: /sensing/camera/camera5/image_rect_color]
      • input/camera5/info [default: /sensing/camera/camera5/camera_info]
      • input/camera5/rois [default: /perception/object_recognition/detection/rois5]
      • input/camera6/image [default: /sensing/camera/camera6/image_rect_color]
      • input/camera6/info [default: /sensing/camera/camera6/camera_info]
      • input/camera6/rois [default: /perception/object_recognition/detection/rois6]
      • input/camera7/image [default: /sensing/camera/camera7/image_rect_color]
      • input/camera7/info [default: /sensing/camera/camera7/camera_info]
      • input/camera7/rois [default: /perception/object_recognition/detection/rois7]
      • input/camera8/image [default: /sensing/camera/camera8/image_rect_color]
      • input/camera8/info [default: /sensing/camera/camera8/camera_info]
      • input/camera8/rois [default: /perception/object_recognition/detection/rois8]
      • input/concatenation_info [default: /sensing/lidar/concatenated/pointcloud_info]
      • input/radar [default: /sensing/radar/detected_objects]
      • input/camera_only/objects [default: /perception/object_recognition/detection/camera_only/objects]
      • input/camera_only/channel [default: camera_streampetr]
      • image_topic_name [default: image_rect_color]
      • segmentation_pointcloud_fusion_camera_ids [default: [0,2,4]]
      • output/tracker_merged_objects [default: /perception/object_recognition/detection/objects]
      • output/tracked_objects [default: /perception/object_recognition/tracking/objects]
      • output/predicted_objects [default: /perception/object_recognition/objects]
      • sync_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/fusion_common.param.yaml]
      • roi_cluster_fusion_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/roi_cluster_fusion.param.yaml]
      • pointpainting_fusion_common_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/pointpainting_common.param.yaml]
      • irregular_object_crop_box_filter_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/crop_box_filter.param.yaml]
      • irregular_object_scan_ground_filter_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/scan_ground_filter.param.yaml]
      • irregular_object_roi_pointcloud_fusion_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/roi_pointcloud_fusion.param.yaml]
      • irregular_object_fusion_camera_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/fusion_cameras.param.yaml]
      • lidar_model_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model]
      • centerpoint_common_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model/centerpoint_common.param.yaml]
      • voxel_grid_based_euclidean_param_path [default: $(var perception_config_dir)/object_recognition/detection/clustering/voxel_grid_based_euclidean_cluster.param.yaml]
      • radar_lanelet_filtering_range_param_path [default: $(var perception_config_dir)/object_recognition/detection/object_filter/radar_lanelet_filter.param.yaml]
      • object_recognition_detection_object_sorter_radar_param_path [default: $(var perception_config_dir)/object_recognition/detection/radar_filter/detected_object_sorter.param.yaml]
      • near_range_camera_vru_param_path [default: $(var perception_config_dir)/object_recognition/detection/camera_vru_detection/near_range_camera_vru_detector.param.yaml]
      • object_recognition_detection_obstacle_pointcloud_based_validator_param_path [default: $(var perception_config_dir)/object_recognition/detection/detected_object_validation/obstacle_pointcloud_based_validator.param.yaml]
      • object_recognition_tracking_multi_object_tracker_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/data_association_matrix.param.yaml]
      • object_recognition_tracking_multi_object_tracker_input_channels_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/input_channels.param.yaml]
      • object_recognition_tracking_multi_object_tracker_node_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/multi_object_tracker_node.param.yaml]
      • euclidean_clustering/param_path [default: $(var perception_config_dir)/object_recognition/detection/euclidean_cluster/label_based_euclidean_cluster.param.yaml]
  • launch/object_recognition/object_recognition_camera_lidar_radar_serial.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • lidar_detection_model [default: centerpoint]
      • use_short_range_detection [default: false]
      • lidar_short_range_detection_model_type [default: centerpoint_short_range]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • bevfusion_model_path [default: $(var data_path)/bevfusion]
      • pointpainting_model_path [default: $(var data_path)/image_projection_based_fusion]
      • camera_2d_detector/model_path [default: $(var data_path)/tensorrt_yolox/yolox-sPlus-opt-pseudoV2-T4-960x960-T4-seg16cls.onnx]
      • camera_2d_detector/label_path [default: $(var data_path)/tensorrt_yolox/label.txt]
      • camera_2d_detector/color_map_path [default: $(var data_path)/tensorrt_yolox/semseg_color_map.csv]
      • camera_2d_detector/roi_remap_path [default: $(var perception_config_dir)/traffic_light_recognition/tensorrt_yolox/roi_label_remap.csv]
      • camera_2d_detector/roi_to_semseg_remap_path [default: $(var perception_config_dir)/traffic_light_recognition/tensorrt_yolox/roi_to_semseg_label_remap.csv]
      • use_object_filter [default: true]
      • objects_filter_method [default: lanelet_filter]
      • use_pointcloud_map [default: true]
      • use_validator [default: false]
      • objects_validation_method [default: obstacle_pointcloud]
      • use_roi_detected_object_validator [default: true]
      • use_detection_by_tracker [default: true]
      • use_low_height_cropbox [default: false]
      • use_low_intensity_cluster_filter [default: true]
      • use_image_segmentation_based_filter [default: false]
      • use_irregular_object_detector [default: true]
      • ml_camera_lidar_merger_priority_mode [default: 0]
      • use_vector_map [default: true]
      • number_of_cameras [default: 6]
      • pointcloud_container_name [default: pointcloud_container]
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • input/obstacle_segmentation/pointcloud [default: /perception/obstacle_segmentation/pointcloud]
      • input/camera0/image [default: /sensing/camera/camera0/image_rect_color]
      • input/camera0/info [default: /sensing/camera/camera0/camera_info]
      • input/camera0/rois [default: /perception/object_recognition/detection/rois0]
      • input/camera1/image [default: /sensing/camera/camera1/image_rect_color]
      • input/camera1/info [default: /sensing/camera/camera1/camera_info]
      • input/camera1/rois [default: /perception/object_recognition/detection/rois1]
      • input/camera2/image [default: /sensing/camera/camera2/image_rect_color]
      • input/camera2/info [default: /sensing/camera/camera2/camera_info]
      • input/camera2/rois [default: /perception/object_recognition/detection/rois2]
      • input/camera3/image [default: /sensing/camera/camera3/image_rect_color]
      • input/camera3/info [default: /sensing/camera/camera3/camera_info]
      • input/camera3/rois [default: /perception/object_recognition/detection/rois3]
      • input/camera4/image [default: /sensing/camera/camera4/image_rect_color]
      • input/camera4/info [default: /sensing/camera/camera4/camera_info]
      • input/camera4/rois [default: /perception/object_recognition/detection/rois4]
      • input/camera5/image [default: /sensing/camera/camera5/image_rect_color]
      • input/camera5/info [default: /sensing/camera/camera5/camera_info]
      • input/camera5/rois [default: /perception/object_recognition/detection/rois5]
      • input/camera6/image [default: /sensing/camera/camera6/image_rect_color]
      • input/camera6/info [default: /sensing/camera/camera6/camera_info]
      • input/camera6/rois [default: /perception/object_recognition/detection/rois6]
      • input/camera7/image [default: /sensing/camera/camera7/image_rect_color]
      • input/camera7/info [default: /sensing/camera/camera7/camera_info]
      • input/camera7/rois [default: /perception/object_recognition/detection/rois7]
      • input/camera8/image [default: /sensing/camera/camera8/image_rect_color]
      • input/camera8/info [default: /sensing/camera/camera8/camera_info]
      • input/camera8/rois [default: /perception/object_recognition/detection/rois8]
      • input/concatenation_info [default: /sensing/lidar/concatenated/pointcloud_info]
      • input/radar/tracked_objects [default: /sensing/radar/tracked_objects]
      • image_topic_name [default: image_rect_color]
      • segmentation_pointcloud_fusion_camera_ids [default: [0,2,4]]
      • output/tracked_objects [default: /perception/object_recognition/tracking/objects]
      • output/predicted_objects [default: /perception/object_recognition/objects]
      • sync_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/fusion_common.param.yaml]
      • roi_cluster_fusion_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/roi_cluster_fusion.param.yaml]
      • pointpainting_fusion_common_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/pointpainting_common.param.yaml]
      • irregular_object_crop_box_filter_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/crop_box_filter.param.yaml]
      • irregular_object_scan_ground_filter_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/scan_ground_filter.param.yaml]
      • irregular_object_roi_pointcloud_fusion_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/roi_pointcloud_fusion.param.yaml]
      • irregular_object_fusion_camera_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/fusion_cameras.param.yaml]
      • lidar_model_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model]
      • centerpoint_common_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model/centerpoint_common.param.yaml]
      • voxel_grid_based_euclidean_param_path [default: $(var perception_config_dir)/object_recognition/detection/clustering/voxel_grid_based_euclidean_cluster.param.yaml]
      • detection_by_tracker_param_path [default: $(var perception_config_dir)/object_recognition/detection/detection_by_tracker/detection_by_tracker.param.yaml]
      • object_recognition_detection_obstacle_pointcloud_based_validator_param_path [default: $(var perception_config_dir)/object_recognition/detection/detected_object_validation/obstacle_pointcloud_based_validator.param.yaml]
      • object_recognition_detection_object_merger_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/detection/object_merger/data_association_matrix.param.yaml]
      • object_recognition_detection_object_merger_distance_threshold_list_path [default: $(var perception_config_dir)/object_recognition/detection/object_merger/overlapped_judge.param.yaml]
      • ml_camera_lidar_object_association_merger_param_path [default: $(var perception_config_dir)/object_recognition/detection/object_merger/ml_camera_lidar_object_association_merger.param.yaml]
      • roi_detected_object_fusion_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/roi_detected_object_fusion.param.yaml]
      • object_recognition_tracking_multi_object_tracker_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/data_association_matrix.param.yaml]
      • object_recognition_tracking_multi_object_tracker_input_channels_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/input_channels.param.yaml]
      • object_recognition_tracking_multi_object_tracker_node_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/multi_object_tracker_node.param.yaml]
      • object_recognition_tracking_radar_tracked_object_sorter_param_path [default: $(var perception_config_dir)/object_recognition/tracking/radar_object_tracker/tracked_object_sorter.param.yaml]
      • object_recognition_tracking_radar_tracked_object_lanelet_filter_param_path [default: $(var perception_config_dir)/object_recognition/tracking/radar_object_tracker/tracked_object_lanelet_filter.param.yaml]
      • object_recognition_tracking_object_merger_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/tracking/tracking_object_merger/data_association_matrix.param.yaml]
      • object_recognition_tracking_object_merger_node_param_path [default: $(var perception_config_dir)/object_recognition/tracking/tracking_object_merger/decorative_tracker_merger.param.yaml]
      • euclidean_clustering/shape_policy [default: 0]
  • launch/object_recognition/object_recognition_lidar.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • lidar_detection_model [default: centerpoint]
      • use_short_range_detection [default: false]
      • lidar_short_range_detection_model_type [default: centerpoint_short_range]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • bevfusion_model_path [default: $(var data_path)/bevfusion]
      • use_pointcloud_map [default: true]
      • use_validator [default: false]
      • objects_validation_method [default: obstacle_pointcloud]
      • use_object_filter [default: true]
      • objects_filter_method [default: lanelet_filter]
      • use_low_height_cropbox [default: false]
      • use_vector_map [default: true]
      • tracker_publish_merged_objects [default: false]
      • pointcloud_container_name [default: pointcloud_container]
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • input/obstacle_segmentation/pointcloud [default: /perception/obstacle_segmentation/pointcloud]
      • output/tracker_merged_objects [default: /perception/object_recognition/detection/objects]
      • output/tracked_objects [default: /perception/object_recognition/tracking/objects]
      • output/predicted_objects [default: /perception/object_recognition/objects]
      • lidar_model_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model]
      • centerpoint_common_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model/centerpoint_common.param.yaml]
      • voxel_grid_based_euclidean_param_path [default: $(var perception_config_dir)/object_recognition/detection/clustering/voxel_grid_based_euclidean_cluster.param.yaml]
      • object_recognition_detection_obstacle_pointcloud_based_validator_param_path [default: $(var perception_config_dir)/object_recognition/detection/detected_object_validation/obstacle_pointcloud_based_validator.param.yaml]
      • object_recognition_tracking_multi_object_tracker_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/data_association_matrix.param.yaml]
      • object_recognition_tracking_multi_object_tracker_input_channels_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/input_channels.param.yaml]
      • object_recognition_tracking_multi_object_tracker_node_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/multi_object_tracker_node.param.yaml]
  • launch/object_recognition/object_recognition_lidar_ptv3.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • lidar_detection_model [default: centerpoint]
      • use_short_range_detection [default: false]
      • lidar_short_range_detection_model_type [default: centerpoint_short_range]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • bevfusion_model_path [default: $(var data_path)/bevfusion]
      • use_vector_map [default: true]
      • tracker_publish_merged_objects [default: false]
      • pointcloud_container_name [default: pointcloud_container]
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • input/semantic_segmentation/pointcloud [default: /perception/segmented/pointcloud]
      • output/tracker_merged_objects [default: /perception/object_recognition/detection/objects]
      • output/tracked_objects [default: /perception/object_recognition/tracking/objects]
      • output/predicted_objects [default: /perception/object_recognition/objects]
      • lidar_model_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model]
      • centerpoint_common_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model/centerpoint_common.param.yaml]
      • object_recognition_tracking_multi_object_tracker_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/data_association_matrix.param.yaml]
      • object_recognition_tracking_multi_object_tracker_input_channels_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/input_channels.param.yaml]
      • object_recognition_tracking_multi_object_tracker_node_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/multi_object_tracker_node.param.yaml]
      • euclidean_clustering/param_path [default: $(var perception_config_dir)/object_recognition/detection/euclidean_cluster/label_based_euclidean_cluster.param.yaml]
  • launch/object_recognition/object_recognition_lidar_radar.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • lidar_detection_model [default: centerpoint]
      • use_short_range_detection [default: false]
      • lidar_short_range_detection_model_type [default: centerpoint_short_range]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • bevfusion_model_path [default: $(var data_path)/bevfusion]
      • use_low_height_cropbox [default: false]
      • use_vector_map [default: true]
      • tracker_publish_merged_objects [default: false]
      • pointcloud_container_name [default: pointcloud_container]
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • input/semantic_segmentation/pointcloud [default: /perception/segmented/pointcloud]
      • input/radar [default: /sensing/radar/detected_objects]
      • output/tracker_merged_objects [default: /perception/object_recognition/detection/objects]
      • output/tracked_objects [default: /perception/object_recognition/tracking/objects]
      • output/predicted_objects [default: /perception/object_recognition/objects]
      • lidar_model_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model]
      • centerpoint_common_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model/centerpoint_common.param.yaml]
      • voxel_grid_based_euclidean_param_path [default: $(var perception_config_dir)/object_recognition/detection/clustering/voxel_grid_based_euclidean_cluster.param.yaml]
      • radar_lanelet_filtering_range_param_path [default: $(var perception_config_dir)/object_recognition/detection/object_filter/radar_lanelet_filter.param.yaml]
      • object_recognition_detection_object_sorter_radar_param_path [default: $(var perception_config_dir)/object_recognition/detection/radar_filter/detected_object_sorter.param.yaml]
      • object_recognition_tracking_multi_object_tracker_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/data_association_matrix.param.yaml]
      • object_recognition_tracking_multi_object_tracker_input_channels_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/input_channels.param.yaml]
      • object_recognition_tracking_multi_object_tracker_node_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/multi_object_tracker_node.param.yaml]
      • euclidean_clustering/param_path [default: $(var perception_config_dir)/object_recognition/detection/euclidean_cluster/label_based_euclidean_cluster.param.yaml]
  • launch/object_recognition/prediction/prediction.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • use_vector_map [default: false]
      • prediction_model_type [default: map_based]
      • input/objects [default: /perception/object_recognition/tracking/objects]
      • output/objects [default: /perception/object_recognition/objects]
      • object_recognition_prediction_map_based_prediction_param_path [default: $(var perception_config_dir)/object_recognition/prediction/map_based_prediction.param.yaml]
  • launch/obstacle_segmentation/ground_segmentation_additional_lidars.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • target_container
      • lidar_name
      • base_frame [default: base_link]
      • use_intra_process [default: true]
      • input/main/pointcloud
      • input/lidar/pointcloud [default: /sensing/lidar/$(var lidar_name)/pointcloud]
      • output/pointcloud
      • output/pointcloud_info [default: $(var output/pointcloud)_info]
      • crop_box_margin_max_z [default: 2.5]
      • crop_box_margin_min_z [default: -2.5]
      • crop_box_filter_node_param_path [default: $(var perception_config_dir)/obstacle_segmentation/ground_segmentation/crop_box_filter.param.yaml]
      • scan_ground_filter_node_param_path [default: $(var perception_config_dir)/obstacle_segmentation/ground_segmentation/scan_ground_filter.param.yaml]
  • launch/obstacle_segmentation/ground_segmentation_ml.launch.xml
      • pointcloud_container_name [default: pointcloud_container]
      • use_pointcloud_container [default: true]
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • output/pointcloud [default: /perception/obstacle_segmentation/pointcloud]
      • output/pointcloud/segmentation [default: /perception/segmented/pointcloud]
      • output/pointcloud/visualization [default: /perception/segmented/pointcloud/visualization]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • model_variant [default: ptv3]
      • model_path [default: $(var data_path)/ptv3]
  • launch/obstacle_segmentation/ground_segmentation_ransac.launch.xml
      • target_container
      • base_frame [default: base_link]
      • use_intra_process [default: true]
      • input_topics
      • input/main/pointcloud
      • output/pointcloud
      • output/pointcloud_info [default: $(var output/pointcloud)_info]
      • detection_area_crop_box_filter_node_param_path
      • ransac_ground_filter_node_param_path [default: $(find-pkg-share autoware_ground_segmentation)/config/ransac_ground_filter.param.yaml]
  • launch/obstacle_segmentation/ground_segmentation_rule_based.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • target_container
      • base_frame [default: base_link]
      • use_intra_process [default: true]
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • output/pointcloud [default: /perception/obstacle_segmentation/pointcloud]
      • crop_box_margin_max_z [default: 2.5]
      • crop_box_margin_min_z [default: -2.5]
      • crop_box_filter_node_param_path [default: $(var perception_config_dir)/obstacle_segmentation/ground_segmentation/crop_box_filter.param.yaml]
      • scan_ground_filter_node_param_path [default: $(var perception_config_dir)/obstacle_segmentation/ground_segmentation/scan_ground_filter.param.yaml]
  • launch/obstacle_segmentation/ground_segmentation_rule_based_cuda.launch.xml
      • target_container
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • output/pointcloud [default: /perception/obstacle_segmentation/pointcloud]
      • output/ground_pointcloud [default: /perception/obstacle_segmentation/ground_pointcloud]
      • cuda_scan_ground_segmentation_filter_node_param_path [default: $(find-pkg-share autoware_ground_segmentation_cuda)/config/cuda_scan_ground_segmentation_filter.param.yaml]
  • launch/occupancy_grid_map/probabilistic_occupancy_grid_map.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • input/obstacle_pointcloud [default: concatenated/pointcloud]
      • input/raw_pointcloud [default: no_ground/oneshot/pointcloud]
      • output [default: /perception/occupancy_grid_map/map]
      • use_intra_process [default: false]
      • use_multithread [default: false]
      • pointcloud_container_name [default: pointcloud_container]
      • occupancy_grid_map_method
      • occupancy_grid_map_param_path [default: $(var perception_config_dir)/occupancy_grid_map/$(var occupancy_grid_map_method).param.yaml]
      • occupancy_grid_map_updater
      • occupancy_grid_map_updater_param_path [default: $(var perception_config_dir)/occupancy_grid_map/$(var occupancy_grid_map_updater)_updater.param.yaml]
      • input_obstacle_pointcloud [default: false]
      • input_obstacle_and_raw_pointcloud [default: true]
      • use_pointcloud_container [default: true]
  • launch/traffic_light_recognition/traffic_light_detector_fine.launch.xml
    • Fine-detection chain for one traffic light camera: map based detector proposes rough ROIs, the fine detector refines them, classifiers label them.
      • perception_config_pkg [default: autoware_perception_config]
      • camera_namespace
      • enable_image_decompressor [default: true]
      • input/vector_map [default: /map/vector_map]
      • input/route [default: /planning/mission_planning/route]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • map_based_detector_param_path [default: $(find-pkg-share autoware_traffic_light_map_based_detector)/config/traffic_light_map_based_detector.param.yaml]
      • fine_detector_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_fine_detector/traffic_light_fine_detector.param.yaml]
      • car_classifier_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_classifier/car_traffic_light_classifier.param.yaml]
      • pedestrian_classifier_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_classifier/pedestrian_traffic_light_classifier.param.yaml]
      • lamp_recognizer_ml_param_path [default: $(var data_path)/traffic_light_classifier/lamp_recognizer_ml.param.yaml]
      • roi_visualizer_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_visualization/traffic_light_roi_visualizer.param.yaml]
      • fine_detector_model_path [default: $(var data_path)/traffic_light_fine_detector/tlr_car_ped_yolox_s_batch_6.onnx]
      • fine_detector_label_path [default: $(var data_path)/traffic_light_fine_detector/tlr_labels.txt]
      • car_classifier_model_path [default: $(var data_path)/traffic_light_classifier/traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • car_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels.txt]
      • pedestrian_classifier_model_path [default: $(var data_path)/traffic_light_classifier/ped_traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • pedestrian_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels_ped.txt]
      • car_classifier_type [default: 1]
      • pedestrian_classifier_type [default: 1]
  • launch/traffic_light_recognition/traffic_light_detector_rough.launch.xml
    • Rough-detection chain for one traffic light camera: map based detector ROIs feed the classifiers directly, without a high accuracy detector.
      • perception_config_pkg [default: autoware_perception_config]
      • camera_namespace
      • enable_image_decompressor [default: true]
      • input/vector_map [default: /map/vector_map]
      • input/route [default: /planning/mission_planning/route]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • map_based_detector_param_path [default: $(find-pkg-share autoware_traffic_light_map_based_detector)/config/traffic_light_map_based_detector.param.yaml]
      • car_classifier_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_classifier/car_traffic_light_classifier.param.yaml]
      • pedestrian_classifier_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_classifier/pedestrian_traffic_light_classifier.param.yaml]
      • lamp_recognizer_ml_param_path [default: $(var data_path)/traffic_light_classifier/lamp_recognizer_ml.param.yaml]
      • roi_visualizer_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_visualization/traffic_light_roi_visualizer.param.yaml]
      • car_classifier_model_path [default: $(var data_path)/traffic_light_classifier/traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • car_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels.txt]
      • pedestrian_classifier_model_path [default: $(var data_path)/traffic_light_classifier/ped_traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • pedestrian_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels_ped.txt]
      • car_classifier_type [default: 1]
      • pedestrian_classifier_type [default: 1]
  • launch/traffic_light_recognition/traffic_light_detector_whole_image.launch.xml
    • Whole-image detection chain for one traffic light camera: yolox detects over the full image, the selector matches detections to map based ROIs, classifiers label them.
      • perception_config_pkg [default: autoware_perception_config]
      • camera_namespace
      • enable_image_decompressor [default: true]
      • input/vector_map [default: /map/vector_map]
      • input/route [default: /planning/mission_planning/route]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • map_based_detector_param_path [default: $(find-pkg-share autoware_traffic_light_map_based_detector)/config/traffic_light_map_based_detector.param.yaml]
      • whole_image_detector_param_path [default: $(var perception_config_dir)/traffic_light_recognition/tensorrt_yolox/yolox_traffic_light_detector.param.yaml]
      • car_classifier_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_classifier/car_traffic_light_classifier.param.yaml]
      • pedestrian_classifier_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_classifier/pedestrian_traffic_light_classifier.param.yaml]
      • lamp_recognizer_ml_param_path [default: $(var data_path)/traffic_light_classifier/lamp_recognizer_ml.param.yaml]
      • roi_visualizer_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_visualization/traffic_light_roi_visualizer.param.yaml]
      • whole_image_detector_model_path [default: $(var data_path)/tensorrt_yolox/yolox_s_car_ped_tl_detector_960_960_batch_1.onnx]
      • whole_image_detector_label_path [default: $(var data_path)/tensorrt_yolox/car_ped_tl_detector_labels.txt]
      • car_classifier_model_path [default: $(var data_path)/traffic_light_classifier/traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • car_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels.txt]
      • pedestrian_classifier_model_path [default: $(var data_path)/traffic_light_classifier/ped_traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • pedestrian_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels_ped.txt]
      • car_classifier_type [default: 1]
      • pedestrian_classifier_type [default: 1]
  • launch/traffic_light_recognition/traffic_light_fusion.launch.xml
    • Post-process shared by all traffic light cameras: multi camera fusion, V2X arbitration, pedestrian signal estimation and map visualization.
      • perception_config_pkg [default: autoware_perception_config]
      • camera_namespaces [default: [camera6, camera7]]
      • input/vector_map [default: /map/vector_map]
      • external/traffic_signals [default: /perception/traffic_light_recognition/external/traffic_signals]
      • internal/traffic_signals [default: /perception/traffic_light_recognition/internal/traffic_signals]
      • judged/traffic_signals [default: /perception/traffic_light_recognition/judged/traffic_signals]
      • output/traffic_signals [default: /perception/traffic_light_recognition/traffic_signals]
      • multi_camera_fusion_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_multi_camera_fusion/traffic_light_multi_camera_fusion.param.yaml]
      • arbiter_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_arbiter/traffic_light_arbiter.param.yaml]
      • crosswalk_estimator_param_path [default: $(var perception_config_dir)/traffic_light_recognition/crosswalk_traffic_light_estimator/crosswalk_traffic_light_estimator.param.yaml]
  • launch/traffic_light_recognition/traffic_light_occlusion_predictor.launch.xml
    • Occlusion check for one traffic light camera: merges the car and pedestrian signals into the per-camera output, masking signals hidden from the camera.
      • perception_config_pkg [default: autoware_perception_config]
      • camera_namespace
      • input/cloud [default: /sensing/lidar/top/pointcloud_raw_ex]
      • input/vector_map [default: /map/vector_map]
      • occlusion_predictor_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_occlusion_predictor/traffic_light_occlusion_predictor.param.yaml]
  • launch/traffic_light_recognition/traffic_light_recognition_fine.launch.xml
    • Traffic light recognition, fine detection mode: per-camera fine detection with occlusion check, then shared fusion. Two cameras as the representative set; add or remove per-camera blocks to match the vehicle.
      • perception_config_pkg [default: autoware_perception_config]
      • input/vector_map [default: /map/vector_map]
      • input/route [default: /planning/mission_planning/route]
      • input/pointcloud [default: /sensing/lidar/top/pointcloud_raw_ex]
      • external/traffic_signals [default: /perception/traffic_light_recognition/external/traffic_signals]
      • output/traffic_signals [default: /perception/traffic_light_recognition/traffic_signals]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • fine_detector_model_path [default: $(var data_path)/traffic_light_fine_detector/tlr_car_ped_yolox_s_batch_6.onnx]
      • fine_detector_label_path [default: $(var data_path)/traffic_light_fine_detector/tlr_labels.txt]
      • lamp_recognizer_ml_param_path [default: $(var data_path)/traffic_light_classifier/lamp_recognizer_ml.param.yaml]
      • car_classifier_model_path [default: $(var data_path)/traffic_light_classifier/traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • car_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels.txt]
      • pedestrian_classifier_model_path [default: $(var data_path)/traffic_light_classifier/ped_traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • pedestrian_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels_ped.txt]
      • car_classifier_type [default: 1]
      • pedestrian_classifier_type [default: 1]
      • base_map_based_detector_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_map_based_detector/base_traffic_light_map_based_detector.param.yaml]
      • extra_map_based_detector_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_map_based_detector/extra_traffic_light_map_based_detector.param.yaml]
  • launch/traffic_light_recognition/traffic_light_recognition_merge.launch.xml
    • Traffic light recognition, merge mode: detection and classification run on other ECUs; this side runs the per-camera occlusion check and the shared fusion. Two cameras as the representative set; add or remove per-camera blocks to match the vehicle.
      • input/vector_map [default: /map/vector_map]
      • input/pointcloud [default: /sensing/lidar/top/pointcloud_raw_ex]
      • external/traffic_signals [default: /perception/traffic_light_recognition/external/traffic_signals]
      • output/traffic_signals [default: /perception/traffic_light_recognition/traffic_signals]
  • launch/traffic_light_recognition/traffic_light_recognition_rough.launch.xml
    • Traffic light recognition, rough detection mode: per-camera map based detection with occlusion check, then shared fusion. One camera as the representative set; add or remove per-camera blocks to match the vehicle.
      • perception_config_pkg [default: autoware_perception_config]
      • input/vector_map [default: /map/vector_map]
      • input/route [default: /planning/mission_planning/route]
      • input/pointcloud [default: /sensing/lidar/top/pointcloud_raw_ex]
      • external/traffic_signals [default: /perception/traffic_light_recognition/external/traffic_signals]
      • output/traffic_signals [default: /perception/traffic_light_recognition/traffic_signals]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • lamp_recognizer_ml_param_path [default: $(var data_path)/traffic_light_classifier/lamp_recognizer_ml.param.yaml]
      • car_classifier_model_path [default: $(var data_path)/traffic_light_classifier/traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • car_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels.txt]
      • pedestrian_classifier_model_path [default: $(var data_path)/traffic_light_classifier/ped_traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • pedestrian_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels_ped.txt]
      • car_classifier_type [default: 1]
      • pedestrian_classifier_type [default: 1]
      • base_map_based_detector_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_map_based_detector/base_traffic_light_map_based_detector.param.yaml]
  • launch/traffic_light_recognition/traffic_light_recognition_whole_image.launch.xml
    • Traffic light recognition, whole-image detection mode: per-camera yolox detection, then shared fusion. Two cameras as the representative set; add or remove per-camera blocks to match the vehicle.
      • perception_config_pkg [default: autoware_perception_config]
      • input/vector_map [default: /map/vector_map]
      • input/route [default: /planning/mission_planning/route]
      • external/traffic_signals [default: /perception/traffic_light_recognition/external/traffic_signals]
      • output/traffic_signals [default: /perception/traffic_light_recognition/traffic_signals]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • whole_image_detector_model_path [default: $(var data_path)/tensorrt_yolox/yolox_s_car_ped_tl_detector_960_960_batch_1.onnx]
      • whole_image_detector_label_path [default: $(var data_path)/tensorrt_yolox/car_ped_tl_detector_labels.txt]
      • lamp_recognizer_ml_param_path [default: $(var data_path)/traffic_light_classifier/lamp_recognizer_ml.param.yaml]
      • car_classifier_model_path [default: $(var data_path)/traffic_light_classifier/traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • car_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels.txt]
      • pedestrian_classifier_model_path [default: $(var data_path)/traffic_light_classifier/ped_traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • pedestrian_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels_ped.txt]
      • car_classifier_type [default: 1]
      • pedestrian_classifier_type [default: 1]
      • base_map_based_detector_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_map_based_detector/base_traffic_light_map_based_detector.param.yaml]
      • extra_map_based_detector_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_map_based_detector/extra_traffic_light_map_based_detector.param.yaml]

Messages

No message files found.

Services

No service files found

Plugins

No plugins found.

Recent questions tagged autoware_perception_launch at Robotics Stack Exchange

No version for distro rolling showing github. Known supported distros are highlighted in the buttons above.
Package symbol

autoware_perception_launch package from autoware_launch repo

autoware_launch autoware_sample_designs autoware_control_config autoware_control_launch autoware_localization_config autoware_localization_launch autoware_map_config autoware_map_launch autoware_perception_config autoware_perception_launch autoware_planning_config autoware_planning_launch autoware_sensing_launch autoware_sensor_driver_launch awsim_labs_sensor_kit_description awsim_labs_sensor_kit_launch awsim_sensor_kit_description awsim_sensor_kit_launch carla_sensor_kit_description carla_sensor_kit_launch sample_sensor_kit_description sample_sensor_kit_launch single_lidar_sensor_kit_description single_lidar_sensor_kit_launch tier4_autoware_api_launch tier4_simulator_launch tier4_system_launch tier4_vehicle_launch awsim_labs_vehicle_description awsim_labs_vehicle_launch sample_vehicle_description sample_vehicle_launch

ROS Distro
github

API Docs
Browse Code

Package Summary

Version 0.53.0
License Apache License 2.0
Build type AMENT_CMAKE
Use RECOMMENDED

Repository Summary

Description
Checkout URI https://github.com/autowarefoundation/autoware_launch.git
VCS Type git
VCS Version main
Last Updated 2026-10-26
Dev Status UNKNOWN
Released UNRELEASED
Contributing Help Wanted (-)
Good First Issues (-)
Pull Requests to Review (-)

Package Description

The autoware_perception_launch package

Maintainers

  • Yukihiro Saito
  • Dai Nguyen
  • Taekjin Lee
  • Yoshi Ri
  • Masato Saeki

Authors

No additional authors.
autoware_universe_launch/autoware_perception_launch/README.md

autoware_perception_launch

Shared perception launch library. The system component launchers live in the product layer (autoware_launch/launch/components/component_perception.launch.xml for the lidar pipeline and component_traffic_light.launch.xml for traffic light recognition); this package hosts the perception feature launchers they combine:

launch/
├── common/                        # shared preprocess feature launchers (pointcloud downsample)
├── object_recognition/
│   ├── object_recognition_<mode>.launch.xml  # detection + tracking + prediction per sensor mode
│   ├── detection/
│   │   ├── detector/              # detectors
│   │   ├── filter/                # filters
│   │   └── merger/                # detection-side serial merger (used by the camera_lidar_radar_serial launcher)
│   └── prediction/
├── obstacle_segmentation/
├── occupancy_grid_map/
└── traffic_light_recognition/
design/module/                     # Autoware System Designer modules; same subtree as launch/
├── object_recognition/
├── obstacle_segmentation/
├── occupancy_grid_map/
└── traffic_light_recognition/

Launch files

  • object_recognition/object_recognition_<mode>.launch.xml — one feature launcher per sensor mode (camera, lidar, lidar_ptv3, lidar_radar, camera_lidar_radar, camera_lidar_radar_serial); hosts the detection stage inline together with the tracker and prediction, and pushes their module namespaces. Camera and radar input topics default to their absolute sensing paths (/sensing/camera/cameraN/..., /sensing/radar/detected_objects), so callers override only what differs. Most modes wire the multi-channel tracker to the detector outputs; camera_lidar_radar_serial instead merges the detectors in the detection stage (optionally regenerating detections through detection-by-tracker), tracks the single merged detected_objects channel, and joins radar tracked objects at the decorative tracker merger. camera runs camera-only detectors (BEV and camera VRU).
  • The clustering backend is file selection, not a flag: lidar runs rule-based euclidean clustering off the obstacle segmentation cloud, while lidar_ptv3 and lidar_radar cluster the PTv3 semantic-segmentation cloud and drop the pointcloud map filter and the object validator that only the rule-based pipeline uses. Both feed their cluster objects to the tracker on the lidar_clustering channel.
  • object_recognition/detection/{detector,filter}/ — single-purpose launchers wired together by the object-recognition launchers. merger/camera_lidar_merger.launch.xml chains the detection-side object mergers for the camera_lidar_radar_serial launcher.
  • obstacle_segmentation/ground_segmentation_<backend>.launch.xml — ground removal backends (rule_based, rule_based_cuda, ml); backend selection is file selection. ground_segmentation_{additional_lidars,ransac}.launch.xml fold extra lidars into the obstacle stream for kits that carry them.
  • occupancy_grid_map/, common/, traffic_light_recognition/ — feature launchers included directly by the system components.

Namespace and container addressing

Callers own the namespaces: the system component pushes perception and object_recognition; object_recognition_<mode>.launch.xml pushes detection/tracking/prediction. The object-recognition launchers address their internal topics through the fixed /perception/object_recognition namespace. Composable nodes load into the shared pointcloud container, whose name is exposed as pointcloud_container_name and aliased to the node/pointcloud_container argument of the detector launchers.

Config

Parameter files are resolved from a config/ tree addressed through the perception_config_pkg argument, which defaults to autoware_perception_config. A product can relocate the whole tree by setting perception_config_pkg once at its entry point — the launch configuration propagates through every include — but it must then ship the same config/ layout. Products that differ in only a few files instead override the corresponding *_param_path arguments, which every feature launcher declares with a default:

<include file="$(find-pkg-share autoware_perception_launch)/launch/object_recognition/object_recognition_lidar.launch.xml">
  <arg name="perception_config_pkg" value="my_perception_config"/>
  <arg name="FOO_NODE_param_path" value="..."/>
</include>

component_perception.launch.xml, component_traffic_light.launch.xml and planning_simulator/perception_simulation.launch.xml each declare perception_config_pkg, so a product switches the tree at the component it launches.

The launch/ tree is the interface downstream distributions adopt as-is; only package.xml, CMakeLists.txt, README.md, and changelogs stay package-local. Keep changes to this tree folder-merge compatible.

Package Dependencies

Please see <exec_depend> in package.xml.

CHANGELOG

Changelog for package autoware_perception_launch

0.53.0 (2026-10-13)

  • Merge remote-tracking branch 'origin/main' into tmp/bot/bump_version_base

  • feat(autoware_launch): run crop_box_filter standalone when agnocast is enabled (#1990)

    * feat(autoware_launch): run crop_box_filter standalone when agnocast is enabled autoware_universe#13384 moves CropBoxFilterComponent to agnocast_wrapper::Node on an agnocast-only executor, which the component containers do not set up. Start each of the eight crop_box_filter nodes as a standalone process when ENABLE_AGNOCAST=1 and keep loading them into their container otherwise. The node names, namespaces, remaps and parameters are the same in both branches.

    * refactor(autoware_launch): launch crop_box_filter through the autoware_node action ---------

  • feat(autoware_perception_launch): run scan_ground_filter standalone when agnocast is enabled (#1991)

    * feat(autoware_perception_launch): run scan_ground_filter standalone when agnocast is enabled autoware_universe#13385 moves ScanGroundFilterComponent to agnocast_wrapper::Node on an agnocast-only executor, which the component containers do not set up. Start each of the three scan_ground_filter nodes as a standalone process when ENABLE_AGNOCAST=1 and keep loading them into their container otherwise. The node names, namespaces, remaps and parameters are the same in both branches.

    * refactor(autoware_perception_launch): launch scan_ground_filter through the autoware_node action ---------

  • feat(autoware_launch): run voxel_grid_downsample_filter standalone when agnocast is enabled (#1992)

    • feat(autoware_launch): run voxel_grid_downsample_filter standalone when agnocast is enabled
    • style(pre-commit): autofix

    * refactor(autoware_launch): launch voxel_grid_downsample_filter through the autoware_node action ---------Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>

  • feat(autoware_sample_designs): compose the AD API, system and vehicle components from design modules (#1976)

    * fix(control, planning): connect the operation mode state and velocity limit inputs Wires collision_detector's operation_mode_state and remaining_distance_time_calculator's planning_velocity through the module designs; their node designs now declare the fixed topic names as remap targets.

    * fix(planning): connect the internal velocity limit path to the selector Forwards motion_velocity_planner's velocity_limit and clear_velocity_limit up to ScenarioPlanning and feeds them into external_velocity_limit_selector, whose internal inputs were unconnected.

    * fix(control, planning): connect the remaining module inputs to their sources Wires the operation mode transition manager's gate and selector inputs, the stop mode operator's vehicle status, the manual lane change handler's route availability and preferred primitive service, the parking costmap, and the route marker. Exposes the vehicle control mode service, the preferred lane service and the system hazard lights command as component interfaces.

    • fix(planning): follow the hazard lights selector input rename
    • fix(planning): add steering_status subscriber to various planning modules
    • feat(vehicle): add vehicle wrapper and launch file for vehicle interface
    • style(pre-commit): autofix
    • fix(planning): add planning factors publishers and connections for various modules
    • refactor(autoware_sample_designs): compose the AD API from design modules
    • Refactor ADAPI module structure and enhance planning factors
    • Updated AdapiRvizHelpers.module.yaml to include helper prefixes for instances and connections.
    • Modified AutowareSample.system.yaml to reorganize ADAPI components, introducing default_adapi for better structure.
    • Expanded planning factors in Planning.module.yaml, LaneDriving.module.yaml, ScenarioPlanning.module.yaml, and associated modules to include additional factors for improved behavior and motion planning.
    • Enhanced connections in behavior planning and motion planning modules to accommodate new planning factors.
    • Ensured consistency across scenario planning modules by aligning

File truncated at 100 lines see the full file

Package Dependencies

Deps Name
ament_cmake_auto
autoware_cmake
autoware_agnocast_wrapper
autoware_bevfusion
autoware_bytetrack
autoware_compare_map_segmentation
autoware_crosswalk_traffic_light_estimator
autoware_cuda_pointcloud_preprocessor
autoware_detected_object_feature_remover
autoware_detected_object_validation
autoware_detection_by_tracker
autoware_elevation_map_loader
autoware_euclidean_cluster
autoware_ground_segmentation
autoware_ground_segmentation_cuda
autoware_image_object_locator
autoware_image_projection_based_fusion
autoware_image_transport_decompressor
autoware_lidar_apollo_instance_segmentation
autoware_lidar_centerpoint
autoware_lidar_transfusion
autoware_map_based_prediction
autoware_multi_object_tracker
autoware_object_merger
autoware_object_sorter
autoware_occupancy_grid_map_outlier_filter
autoware_perception_config
autoware_pointcloud_preprocessor
autoware_probabilistic_occupancy_grid_map
autoware_ptv3
autoware_raindrop_cluster_filter
autoware_shape_estimation
autoware_simpl_prediction
autoware_simple_object_merger
autoware_tensorrt_bevdet
autoware_tensorrt_yolox
autoware_tracking_object_merger
autoware_traffic_light_arbiter
autoware_traffic_light_category_merger
autoware_traffic_light_classifier
autoware_traffic_light_fine_detector
autoware_traffic_light_map_based_detector
autoware_traffic_light_multi_camera_fusion
autoware_traffic_light_occlusion_predictor
autoware_traffic_light_selector
autoware_traffic_light_visualization
rclcpp_components
topic_tools
ament_lint_auto
autoware_lint_common

System Dependencies

No direct system dependencies.

Dependant Packages

Name Deps
autoware_launch
reaction_analyzer

Launch files

  • launch/common/pointcloud_downsample.launch.xml
      • pointcloud_container_name
      • cuda_pointcloud_preprocessing [default: false]
      • input/pointcloud
      • output/pointcloud
      • common_downsample_voxel_size_x [default: 0.05]
      • common_downsample_voxel_size_y [default: 0.05]
      • common_downsample_voxel_size_z [default: 0.05]
  • launch/object_recognition/detection/detector/camera_2d_detector.launch.xml
      • image_raw0 [default: /sensing/camera/camera0/image_raw]
      • image_raw1 [default: /sensing/camera/camera1/image_raw]
      • image_raw2 [default: /sensing/camera/camera2/image_raw]
      • image_raw3 [default: /sensing/camera/camera3/image_raw]
      • image_raw4 [default: /sensing/camera/camera4/image_raw]
      • image_raw5 [default: /sensing/camera/camera5/image_raw]
      • image_raw6 [default: /sensing/camera/camera6/image_raw]
      • image_raw7 [default: /sensing/camera/camera7/image_raw]
      • image_raw8 [default: /sensing/camera/camera8/image_raw]
      • image_raw9 [default: /sensing/camera/camera9/image_raw]
      • image_number [default: 1]
      • camera_index [default: 0]
      • use_bytetrack [default: true]
      • enable_visualizer [default: false]
      • camera_2d_detector/model_path
      • camera_2d_detector/label_path
      • camera_2d_detector/color_map_path
      • camera_2d_detector/roi_remap_path
      • camera_2d_detector/roi_to_semseg_remap_path
      • tensorrt_yolox_ns [default: ]
  • launch/object_recognition/detection/detector/camera_bev_detector.launch.xml
      • input/camera0/image
      • input/camera0/info
      • input/camera1/image
      • input/camera1/info
      • input/camera2/image
      • input/camera2/info
      • input/camera3/image
      • input/camera3/info
      • input/camera4/image
      • input/camera4/info
      • input/camera5/image
      • input/camera5/info
      • output/objects
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • bevdet_model_name [default: bevdet_one_lt_d]
      • bevdet_model_path [default: $(var data_path)/tensorrt_bevdet]
  • launch/object_recognition/detection/detector/camera_lidar_detector.launch.xml
      • ns
      • lidar_detection_model_type
      • lidar_detection_model_name
      • use_low_intensity_cluster_filter
      • use_image_segmentation_based_filter
      • number_of_cameras
      • input/camera0/image
      • input/camera0/info
      • input/camera0/rois
      • input/camera1/image
      • input/camera1/info
      • input/camera1/rois
      • input/camera2/image
      • input/camera2/info
      • input/camera2/rois
      • input/camera3/image
      • input/camera3/info
      • input/camera3/rois
      • input/camera4/image
      • input/camera4/info
      • input/camera4/rois
      • input/camera5/image
      • input/camera5/info
      • input/camera5/rois
      • input/camera6/image
      • input/camera6/info
      • input/camera6/rois
      • input/camera7/image
      • input/camera7/info
      • input/camera7/rois
      • input/camera8/image
      • input/camera8/info
      • input/camera8/rois
      • input/concatenation_info
      • segmentation_pointcloud_fusion_camera_ids
      • image_topic_name
      • sync_param_path
      • voxel_grid_based_euclidean_param_path
      • lidar_model_param_path
      • pointpainting_model_path [default: $(var data_path)/image_projection_based_fusion]
      • pointpainting_fusion_common_param_path
      • roi_cluster_fusion_param_path
      • node/pointcloud_container
      • input/pointcloud
      • input/pointcloud_map/pointcloud
      • input/obstacle_segmentation/pointcloud
      • output/ml_detector/objects
      • output/rule_detector/objects
      • camera_2d_detector/model_path
      • camera_2d_detector/label_path
      • camera_2d_detector/color_map_path
      • camera_2d_detector/roi_remap_path
      • camera_2d_detector/roi_to_semseg_remap_path
      • enable_2d_detection [default: false]
  • launch/object_recognition/detection/detector/camera_lidar_irregular_object_detector.launch.xml
      • pipeline_ns
      • node/pointcloud_container
      • output/objects [default: objects]
      • input/concatenation_info
      • input/pointcloud
      • irregular_object_crop_box_filter_node_param_path
      • irregular_object_scan_ground_filter_node_param_path
      • irregular_object_roi_pointcloud_fusion_node_param_path
      • irregular_object_fusion_camera_param_path
      • sync_param_path
  • launch/object_recognition/detection/detector/camera_vru_detector.launch.xml
      • input/camera0/info [default: /sensing/camera/camera0/camera_info]
      • input/camera0/rois [default: /perception/object_recognition/detection/rois0]
      • input/camera1/info [default: /sensing/camera/camera1/camera_info]
      • input/camera1/rois [default: /perception/object_recognition/detection/rois1]
      • input/camera2/info [default: /sensing/camera/camera2/camera_info]
      • input/camera2/rois [default: /perception/object_recognition/detection/rois2]
      • input/camera3/info [default: /sensing/camera/camera3/camera_info]
      • input/camera3/rois [default: /perception/object_recognition/detection/rois3]
      • input/camera4/info [default: /sensing/camera/camera4/camera_info]
      • input/camera4/rois [default: /perception/object_recognition/detection/rois4]
      • input/camera5/info [default: /sensing/camera/camera5/camera_info]
      • input/camera5/rois [default: /perception/object_recognition/detection/rois5]
      • input/camera6/info [default: /sensing/camera/camera6/camera_info]
      • input/camera6/rois [default: /perception/object_recognition/detection/rois6]
      • input/camera7/info [default: /sensing/camera/camera7/camera_info]
      • input/camera7/rois [default: /perception/object_recognition/detection/rois7]
      • output/objects [default: /perception/object_recognition/detection/camera_vru/objects]
      • bbox_object_locator_param_path
      • rois_ids [default: [0, 1]]
  • launch/object_recognition/detection/detector/irregular_object_detector.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • target_container
      • base_frame [default: base_link]
      • use_intra_process [default: true]
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • input/concatenation_info [default: /sensing/lidar/concatenated/pointcloud_info]
      • output/clusters
      • crop_box_filter_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/crop_box_filter.param.yaml]
      • scan_ground_filter_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/scan_ground_filter.param.yaml]
      • roi_pointcloud_fusion_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/roi_pointcloud_fusion.param.yaml]
      • fusion_camera_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/fusion_cameras.param.yaml]
      • sync_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/fusion_common.param.yaml]
  • launch/object_recognition/detection/detector/lidar_dnn_detector.launch.xml
      • lidar_detection_model_type
      • lidar_detection_model_name
      • lidar_model_param_path
      • bevfusion_model_path [default: $(var data_path)/bevfusion]
      • centerpoint_model_path [default: $(var data_path)/lidar_centerpoint]
      • centerpoint_common_param_path [default: $(find-pkg-share autoware_lidar_centerpoint)/config/centerpoint_common.param.yaml]
      • transfusion_model_path [default: $(var data_path)/lidar_transfusion]
      • use_short_range_detection [default: false]
      • lidar_short_range_detection_model_type
      • node/pointcloud_container
      • input/pointcloud
      • output/objects
      • output/short_range_objects
  • launch/object_recognition/detection/detector/lidar_rule_detector.launch.xml
      • ns
      • node/pointcloud_container
      • input/pointcloud_map/pointcloud
      • input/obstacle_segmentation/pointcloud
      • output/cluster_objects
      • output/objects
      • voxel_grid_based_euclidean_param_path
      • use_low_height_cropbox [default: false]
  • launch/object_recognition/detection/detector/lidar_semantic_segmentation_detector.launch.xml
      • euclidean_clustering/param_path
      • input/pointcloud
      • output/objects
  • launch/object_recognition/detection/detector/tracker_based_detector.launch.xml
      • detection_by_tracker_param_path
      • input/clusters
      • input/tracked_objects
      • output/objects
  • launch/object_recognition/detection/filter/object_filter.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • objects_filter_method [default: lanelet_filter]
      • object_recognition_detection_object_lanelet_filter_param_path [default: $(var perception_config_dir)/object_recognition/detection/object_filter/object_lanelet_filter.param.yaml]
      • object_recognition_detection_object_position_filter_param_path [default: $(var perception_config_dir)/object_recognition/detection/object_filter/object_position_filter.param.yaml]
      • input/objects
      • output/objects
  • launch/object_recognition/detection/filter/object_validator.launch.xml
      • objects_validation_method
      • object_recognition_detection_obstacle_pointcloud_based_validator_param_path
      • input/obstacle_pointcloud
      • input/objects
      • output/objects
  • launch/object_recognition/detection/filter/pointcloud_downsample_filter.launch.xml
      • target_container
      • use_intra_process [default: true]
      • input/pointcloud
      • output/pointcloud
      • downsample_voxel_size [default: 0.1]
  • launch/object_recognition/detection/filter/pointcloud_map_filter.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • target_container
      • use_intra_process [default: true]
      • input/pointcloud
      • output/pointcloud
      • output/downsampled/pointcloud [default: /perception/obstacle_segmentation/pointcloud_map_filtered/downsampled/pointcloud]
      • downsample_voxel_size [default: 0.1]
      • voxel_based_compare_map_filter_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/pointcloud_filter/voxel_based_compare_map_filter.param.yaml]
  • launch/object_recognition/detection/filter/radar_filter.launch.xml
      • object_sorter_param_path
      • radar_lanelet_filtering_range_param_path
      • input/radar
      • output/objects
  • launch/object_recognition/detection/merger/camera_lidar_merger.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • object_recognition_detection_object_merger_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/detection/object_merger/data_association_matrix.param.yaml]
      • object_recognition_detection_object_merger_distance_threshold_list_path [default: $(var perception_config_dir)/object_recognition/detection/object_merger/overlapped_judge.param.yaml]
      • ml_camera_lidar_object_association_merger_param_path [default: $(var perception_config_dir)/object_recognition/detection/object_merger/ml_camera_lidar_object_association_merger.param.yaml]
      • roi_detected_object_fusion_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/roi_detected_object_fusion.param.yaml]
      • sync_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/fusion_common.param.yaml]
      • lidar_detection_model_type
      • use_detection_by_tracker
      • use_irregular_object_detector
      • use_roi_detected_object_validator
      • euclidean_clustering/shape_policy
      • use_object_filter
      • objects_filter_method
      • number_of_cameras
      • input/camera0/image
      • input/camera0/info
      • input/camera0/rois
      • input/camera1/image
      • input/camera1/info
      • input/camera1/rois
      • input/camera2/image
      • input/camera2/info
      • input/camera2/rois
      • input/camera3/image
      • input/camera3/info
      • input/camera3/rois
      • input/camera4/image
      • input/camera4/info
      • input/camera4/rois
      • input/camera5/image
      • input/camera5/info
      • input/camera5/rois
      • input/camera6/image
      • input/camera6/info
      • input/camera6/rois
      • input/camera7/image
      • input/camera7/info
      • input/camera7/rois
      • input/camera8/image
      • input/camera8/info
      • input/camera8/rois
      • input/concatenation_info [default: /sensing/lidar/concatenated/pointcloud_info]
      • input/lidar_ml/objects
      • input/lidar_cluster/objects
      • input/irregular_object/objects [default: ]
      • input/detection_by_tracker/objects
      • output/objects [default: objects]
      • alpha_merger_priority_mode [default: 0]
  • launch/object_recognition/object_recognition_camera.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • use_camera_vru_detector [default: false]
      • camera_vru_detector_rois_ids [default: [0]]
      • use_camera_bev_detector [default: true]
      • use_vector_map [default: true]
      • tracker_publish_merged_objects [default: false]
      • input/camera0/image [default: /sensing/camera/camera0/image_rect_color]
      • input/camera0/info [default: /sensing/camera/camera0/camera_info]
      • input/camera0/rois [default: /perception/object_recognition/detection/rois0]
      • input/camera1/image [default: /sensing/camera/camera1/image_rect_color]
      • input/camera1/info [default: /sensing/camera/camera1/camera_info]
      • input/camera1/rois [default: /perception/object_recognition/detection/rois1]
      • input/camera2/image [default: /sensing/camera/camera2/image_rect_color]
      • input/camera2/info [default: /sensing/camera/camera2/camera_info]
      • input/camera2/rois [default: /perception/object_recognition/detection/rois2]
      • input/camera3/image [default: /sensing/camera/camera3/image_rect_color]
      • input/camera3/info [default: /sensing/camera/camera3/camera_info]
      • input/camera3/rois [default: /perception/object_recognition/detection/rois3]
      • input/camera4/image [default: /sensing/camera/camera4/image_rect_color]
      • input/camera4/info [default: /sensing/camera/camera4/camera_info]
      • input/camera4/rois [default: /perception/object_recognition/detection/rois4]
      • input/camera5/image [default: /sensing/camera/camera5/image_rect_color]
      • input/camera5/info [default: /sensing/camera/camera5/camera_info]
      • input/camera5/rois [default: /perception/object_recognition/detection/rois5]
      • input/camera6/image [default: /sensing/camera/camera6/image_rect_color]
      • input/camera6/info [default: /sensing/camera/camera6/camera_info]
      • input/camera6/rois [default: /perception/object_recognition/detection/rois6]
      • input/camera7/image [default: /sensing/camera/camera7/image_rect_color]
      • input/camera7/info [default: /sensing/camera/camera7/camera_info]
      • input/camera7/rois [default: /perception/object_recognition/detection/rois7]
      • output/tracker_merged_objects [default: /perception/object_recognition/detection/objects]
      • output/tracked_objects [default: /perception/object_recognition/tracking/objects]
      • output/predicted_objects [default: /perception/object_recognition/objects]
      • near_range_camera_vru_param_path [default: $(var perception_config_dir)/object_recognition/detection/camera_vru_detection/near_range_camera_vru_detector.param.yaml]
      • object_recognition_tracking_multi_object_tracker_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/data_association_matrix.param.yaml]
      • object_recognition_tracking_multi_object_tracker_input_channels_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/input_channels.param.yaml]
      • object_recognition_tracking_multi_object_tracker_node_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/multi_object_tracker_node.param.yaml]
  • launch/object_recognition/object_recognition_camera_lidar_radar.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • lidar_detection_model [default: centerpoint]
      • use_short_range_detection [default: false]
      • lidar_short_range_detection_model_type [default: centerpoint_short_range]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • bevfusion_model_path [default: $(var data_path)/bevfusion]
      • pointpainting_model_path [default: $(var data_path)/image_projection_based_fusion]
      • camera_2d_detector/model_path [default: $(var data_path)/tensorrt_yolox/yolox-sPlus-opt-pseudoV2-T4-960x960-T4-seg16cls.onnx]
      • camera_2d_detector/label_path [default: $(var data_path)/tensorrt_yolox/label.txt]
      • camera_2d_detector/color_map_path [default: $(var data_path)/tensorrt_yolox/semseg_color_map.csv]
      • camera_2d_detector/roi_remap_path [default: $(var perception_config_dir)/traffic_light_recognition/tensorrt_yolox/roi_label_remap.csv]
      • camera_2d_detector/roi_to_semseg_remap_path [default: $(var perception_config_dir)/traffic_light_recognition/tensorrt_yolox/roi_to_semseg_label_remap.csv]
      • use_semantic_segmentation_ptv3 [default: false]
      • use_object_filter [default: true]
      • use_pointcloud_map [default: true]
      • use_validator [default: false]
      • objects_validation_method [default: obstacle_pointcloud]
      • use_low_height_cropbox [default: false]
      • use_low_intensity_cluster_filter [default: true]
      • use_image_segmentation_based_filter [default: false]
      • use_irregular_object_detector [default: true]
      • use_camera_vru_detector [default: false]
      • camera_vru_detector_rois_ids [default: [0]]
      • use_camera_only_detection [default: false]
      • use_vector_map [default: true]
      • tracker_publish_merged_objects [default: false]
      • number_of_cameras [default: 6]
      • pointcloud_container_name [default: pointcloud_container]
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • input/semantic_segmentation/pointcloud [default: /perception/segmented/pointcloud]
      • input/obstacle_segmentation/pointcloud [default: /perception/obstacle_segmentation/pointcloud]
      • input/camera0/image [default: /sensing/camera/camera0/image_rect_color]
      • input/camera0/info [default: /sensing/camera/camera0/camera_info]
      • input/camera0/rois [default: /perception/object_recognition/detection/rois0]
      • input/camera1/image [default: /sensing/camera/camera1/image_rect_color]
      • input/camera1/info [default: /sensing/camera/camera1/camera_info]
      • input/camera1/rois [default: /perception/object_recognition/detection/rois1]
      • input/camera2/image [default: /sensing/camera/camera2/image_rect_color]
      • input/camera2/info [default: /sensing/camera/camera2/camera_info]
      • input/camera2/rois [default: /perception/object_recognition/detection/rois2]
      • input/camera3/image [default: /sensing/camera/camera3/image_rect_color]
      • input/camera3/info [default: /sensing/camera/camera3/camera_info]
      • input/camera3/rois [default: /perception/object_recognition/detection/rois3]
      • input/camera4/image [default: /sensing/camera/camera4/image_rect_color]
      • input/camera4/info [default: /sensing/camera/camera4/camera_info]
      • input/camera4/rois [default: /perception/object_recognition/detection/rois4]
      • input/camera5/image [default: /sensing/camera/camera5/image_rect_color]
      • input/camera5/info [default: /sensing/camera/camera5/camera_info]
      • input/camera5/rois [default: /perception/object_recognition/detection/rois5]
      • input/camera6/image [default: /sensing/camera/camera6/image_rect_color]
      • input/camera6/info [default: /sensing/camera/camera6/camera_info]
      • input/camera6/rois [default: /perception/object_recognition/detection/rois6]
      • input/camera7/image [default: /sensing/camera/camera7/image_rect_color]
      • input/camera7/info [default: /sensing/camera/camera7/camera_info]
      • input/camera7/rois [default: /perception/object_recognition/detection/rois7]
      • input/camera8/image [default: /sensing/camera/camera8/image_rect_color]
      • input/camera8/info [default: /sensing/camera/camera8/camera_info]
      • input/camera8/rois [default: /perception/object_recognition/detection/rois8]
      • input/concatenation_info [default: /sensing/lidar/concatenated/pointcloud_info]
      • input/radar [default: /sensing/radar/detected_objects]
      • input/camera_only/objects [default: /perception/object_recognition/detection/camera_only/objects]
      • input/camera_only/channel [default: camera_streampetr]
      • image_topic_name [default: image_rect_color]
      • segmentation_pointcloud_fusion_camera_ids [default: [0,2,4]]
      • output/tracker_merged_objects [default: /perception/object_recognition/detection/objects]
      • output/tracked_objects [default: /perception/object_recognition/tracking/objects]
      • output/predicted_objects [default: /perception/object_recognition/objects]
      • sync_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/fusion_common.param.yaml]
      • roi_cluster_fusion_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/roi_cluster_fusion.param.yaml]
      • pointpainting_fusion_common_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/pointpainting_common.param.yaml]
      • irregular_object_crop_box_filter_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/crop_box_filter.param.yaml]
      • irregular_object_scan_ground_filter_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/scan_ground_filter.param.yaml]
      • irregular_object_roi_pointcloud_fusion_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/roi_pointcloud_fusion.param.yaml]
      • irregular_object_fusion_camera_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/fusion_cameras.param.yaml]
      • lidar_model_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model]
      • centerpoint_common_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model/centerpoint_common.param.yaml]
      • voxel_grid_based_euclidean_param_path [default: $(var perception_config_dir)/object_recognition/detection/clustering/voxel_grid_based_euclidean_cluster.param.yaml]
      • radar_lanelet_filtering_range_param_path [default: $(var perception_config_dir)/object_recognition/detection/object_filter/radar_lanelet_filter.param.yaml]
      • object_recognition_detection_object_sorter_radar_param_path [default: $(var perception_config_dir)/object_recognition/detection/radar_filter/detected_object_sorter.param.yaml]
      • near_range_camera_vru_param_path [default: $(var perception_config_dir)/object_recognition/detection/camera_vru_detection/near_range_camera_vru_detector.param.yaml]
      • object_recognition_detection_obstacle_pointcloud_based_validator_param_path [default: $(var perception_config_dir)/object_recognition/detection/detected_object_validation/obstacle_pointcloud_based_validator.param.yaml]
      • object_recognition_tracking_multi_object_tracker_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/data_association_matrix.param.yaml]
      • object_recognition_tracking_multi_object_tracker_input_channels_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/input_channels.param.yaml]
      • object_recognition_tracking_multi_object_tracker_node_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/multi_object_tracker_node.param.yaml]
      • euclidean_clustering/param_path [default: $(var perception_config_dir)/object_recognition/detection/euclidean_cluster/label_based_euclidean_cluster.param.yaml]
  • launch/object_recognition/object_recognition_camera_lidar_radar_serial.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • lidar_detection_model [default: centerpoint]
      • use_short_range_detection [default: false]
      • lidar_short_range_detection_model_type [default: centerpoint_short_range]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • bevfusion_model_path [default: $(var data_path)/bevfusion]
      • pointpainting_model_path [default: $(var data_path)/image_projection_based_fusion]
      • camera_2d_detector/model_path [default: $(var data_path)/tensorrt_yolox/yolox-sPlus-opt-pseudoV2-T4-960x960-T4-seg16cls.onnx]
      • camera_2d_detector/label_path [default: $(var data_path)/tensorrt_yolox/label.txt]
      • camera_2d_detector/color_map_path [default: $(var data_path)/tensorrt_yolox/semseg_color_map.csv]
      • camera_2d_detector/roi_remap_path [default: $(var perception_config_dir)/traffic_light_recognition/tensorrt_yolox/roi_label_remap.csv]
      • camera_2d_detector/roi_to_semseg_remap_path [default: $(var perception_config_dir)/traffic_light_recognition/tensorrt_yolox/roi_to_semseg_label_remap.csv]
      • use_object_filter [default: true]
      • objects_filter_method [default: lanelet_filter]
      • use_pointcloud_map [default: true]
      • use_validator [default: false]
      • objects_validation_method [default: obstacle_pointcloud]
      • use_roi_detected_object_validator [default: true]
      • use_detection_by_tracker [default: true]
      • use_low_height_cropbox [default: false]
      • use_low_intensity_cluster_filter [default: true]
      • use_image_segmentation_based_filter [default: false]
      • use_irregular_object_detector [default: true]
      • ml_camera_lidar_merger_priority_mode [default: 0]
      • use_vector_map [default: true]
      • number_of_cameras [default: 6]
      • pointcloud_container_name [default: pointcloud_container]
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • input/obstacle_segmentation/pointcloud [default: /perception/obstacle_segmentation/pointcloud]
      • input/camera0/image [default: /sensing/camera/camera0/image_rect_color]
      • input/camera0/info [default: /sensing/camera/camera0/camera_info]
      • input/camera0/rois [default: /perception/object_recognition/detection/rois0]
      • input/camera1/image [default: /sensing/camera/camera1/image_rect_color]
      • input/camera1/info [default: /sensing/camera/camera1/camera_info]
      • input/camera1/rois [default: /perception/object_recognition/detection/rois1]
      • input/camera2/image [default: /sensing/camera/camera2/image_rect_color]
      • input/camera2/info [default: /sensing/camera/camera2/camera_info]
      • input/camera2/rois [default: /perception/object_recognition/detection/rois2]
      • input/camera3/image [default: /sensing/camera/camera3/image_rect_color]
      • input/camera3/info [default: /sensing/camera/camera3/camera_info]
      • input/camera3/rois [default: /perception/object_recognition/detection/rois3]
      • input/camera4/image [default: /sensing/camera/camera4/image_rect_color]
      • input/camera4/info [default: /sensing/camera/camera4/camera_info]
      • input/camera4/rois [default: /perception/object_recognition/detection/rois4]
      • input/camera5/image [default: /sensing/camera/camera5/image_rect_color]
      • input/camera5/info [default: /sensing/camera/camera5/camera_info]
      • input/camera5/rois [default: /perception/object_recognition/detection/rois5]
      • input/camera6/image [default: /sensing/camera/camera6/image_rect_color]
      • input/camera6/info [default: /sensing/camera/camera6/camera_info]
      • input/camera6/rois [default: /perception/object_recognition/detection/rois6]
      • input/camera7/image [default: /sensing/camera/camera7/image_rect_color]
      • input/camera7/info [default: /sensing/camera/camera7/camera_info]
      • input/camera7/rois [default: /perception/object_recognition/detection/rois7]
      • input/camera8/image [default: /sensing/camera/camera8/image_rect_color]
      • input/camera8/info [default: /sensing/camera/camera8/camera_info]
      • input/camera8/rois [default: /perception/object_recognition/detection/rois8]
      • input/concatenation_info [default: /sensing/lidar/concatenated/pointcloud_info]
      • input/radar/tracked_objects [default: /sensing/radar/tracked_objects]
      • image_topic_name [default: image_rect_color]
      • segmentation_pointcloud_fusion_camera_ids [default: [0,2,4]]
      • output/tracked_objects [default: /perception/object_recognition/tracking/objects]
      • output/predicted_objects [default: /perception/object_recognition/objects]
      • sync_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/fusion_common.param.yaml]
      • roi_cluster_fusion_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/roi_cluster_fusion.param.yaml]
      • pointpainting_fusion_common_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/pointpainting_common.param.yaml]
      • irregular_object_crop_box_filter_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/crop_box_filter.param.yaml]
      • irregular_object_scan_ground_filter_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/scan_ground_filter.param.yaml]
      • irregular_object_roi_pointcloud_fusion_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/roi_pointcloud_fusion.param.yaml]
      • irregular_object_fusion_camera_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/fusion_cameras.param.yaml]
      • lidar_model_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model]
      • centerpoint_common_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model/centerpoint_common.param.yaml]
      • voxel_grid_based_euclidean_param_path [default: $(var perception_config_dir)/object_recognition/detection/clustering/voxel_grid_based_euclidean_cluster.param.yaml]
      • detection_by_tracker_param_path [default: $(var perception_config_dir)/object_recognition/detection/detection_by_tracker/detection_by_tracker.param.yaml]
      • object_recognition_detection_obstacle_pointcloud_based_validator_param_path [default: $(var perception_config_dir)/object_recognition/detection/detected_object_validation/obstacle_pointcloud_based_validator.param.yaml]
      • object_recognition_detection_object_merger_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/detection/object_merger/data_association_matrix.param.yaml]
      • object_recognition_detection_object_merger_distance_threshold_list_path [default: $(var perception_config_dir)/object_recognition/detection/object_merger/overlapped_judge.param.yaml]
      • ml_camera_lidar_object_association_merger_param_path [default: $(var perception_config_dir)/object_recognition/detection/object_merger/ml_camera_lidar_object_association_merger.param.yaml]
      • roi_detected_object_fusion_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/roi_detected_object_fusion.param.yaml]
      • object_recognition_tracking_multi_object_tracker_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/data_association_matrix.param.yaml]
      • object_recognition_tracking_multi_object_tracker_input_channels_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/input_channels.param.yaml]
      • object_recognition_tracking_multi_object_tracker_node_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/multi_object_tracker_node.param.yaml]
      • object_recognition_tracking_radar_tracked_object_sorter_param_path [default: $(var perception_config_dir)/object_recognition/tracking/radar_object_tracker/tracked_object_sorter.param.yaml]
      • object_recognition_tracking_radar_tracked_object_lanelet_filter_param_path [default: $(var perception_config_dir)/object_recognition/tracking/radar_object_tracker/tracked_object_lanelet_filter.param.yaml]
      • object_recognition_tracking_object_merger_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/tracking/tracking_object_merger/data_association_matrix.param.yaml]
      • object_recognition_tracking_object_merger_node_param_path [default: $(var perception_config_dir)/object_recognition/tracking/tracking_object_merger/decorative_tracker_merger.param.yaml]
      • euclidean_clustering/shape_policy [default: 0]
  • launch/object_recognition/object_recognition_lidar.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • lidar_detection_model [default: centerpoint]
      • use_short_range_detection [default: false]
      • lidar_short_range_detection_model_type [default: centerpoint_short_range]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • bevfusion_model_path [default: $(var data_path)/bevfusion]
      • use_pointcloud_map [default: true]
      • use_validator [default: false]
      • objects_validation_method [default: obstacle_pointcloud]
      • use_object_filter [default: true]
      • objects_filter_method [default: lanelet_filter]
      • use_low_height_cropbox [default: false]
      • use_vector_map [default: true]
      • tracker_publish_merged_objects [default: false]
      • pointcloud_container_name [default: pointcloud_container]
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • input/obstacle_segmentation/pointcloud [default: /perception/obstacle_segmentation/pointcloud]
      • output/tracker_merged_objects [default: /perception/object_recognition/detection/objects]
      • output/tracked_objects [default: /perception/object_recognition/tracking/objects]
      • output/predicted_objects [default: /perception/object_recognition/objects]
      • lidar_model_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model]
      • centerpoint_common_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model/centerpoint_common.param.yaml]
      • voxel_grid_based_euclidean_param_path [default: $(var perception_config_dir)/object_recognition/detection/clustering/voxel_grid_based_euclidean_cluster.param.yaml]
      • object_recognition_detection_obstacle_pointcloud_based_validator_param_path [default: $(var perception_config_dir)/object_recognition/detection/detected_object_validation/obstacle_pointcloud_based_validator.param.yaml]
      • object_recognition_tracking_multi_object_tracker_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/data_association_matrix.param.yaml]
      • object_recognition_tracking_multi_object_tracker_input_channels_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/input_channels.param.yaml]
      • object_recognition_tracking_multi_object_tracker_node_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/multi_object_tracker_node.param.yaml]
  • launch/object_recognition/object_recognition_lidar_ptv3.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • lidar_detection_model [default: centerpoint]
      • use_short_range_detection [default: false]
      • lidar_short_range_detection_model_type [default: centerpoint_short_range]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • bevfusion_model_path [default: $(var data_path)/bevfusion]
      • use_vector_map [default: true]
      • tracker_publish_merged_objects [default: false]
      • pointcloud_container_name [default: pointcloud_container]
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • input/semantic_segmentation/pointcloud [default: /perception/segmented/pointcloud]
      • output/tracker_merged_objects [default: /perception/object_recognition/detection/objects]
      • output/tracked_objects [default: /perception/object_recognition/tracking/objects]
      • output/predicted_objects [default: /perception/object_recognition/objects]
      • lidar_model_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model]
      • centerpoint_common_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model/centerpoint_common.param.yaml]
      • object_recognition_tracking_multi_object_tracker_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/data_association_matrix.param.yaml]
      • object_recognition_tracking_multi_object_tracker_input_channels_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/input_channels.param.yaml]
      • object_recognition_tracking_multi_object_tracker_node_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/multi_object_tracker_node.param.yaml]
      • euclidean_clustering/param_path [default: $(var perception_config_dir)/object_recognition/detection/euclidean_cluster/label_based_euclidean_cluster.param.yaml]
  • launch/object_recognition/object_recognition_lidar_radar.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • lidar_detection_model [default: centerpoint]
      • use_short_range_detection [default: false]
      • lidar_short_range_detection_model_type [default: centerpoint_short_range]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • bevfusion_model_path [default: $(var data_path)/bevfusion]
      • use_low_height_cropbox [default: false]
      • use_vector_map [default: true]
      • tracker_publish_merged_objects [default: false]
      • pointcloud_container_name [default: pointcloud_container]
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • input/semantic_segmentation/pointcloud [default: /perception/segmented/pointcloud]
      • input/radar [default: /sensing/radar/detected_objects]
      • output/tracker_merged_objects [default: /perception/object_recognition/detection/objects]
      • output/tracked_objects [default: /perception/object_recognition/tracking/objects]
      • output/predicted_objects [default: /perception/object_recognition/objects]
      • lidar_model_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model]
      • centerpoint_common_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model/centerpoint_common.param.yaml]
      • voxel_grid_based_euclidean_param_path [default: $(var perception_config_dir)/object_recognition/detection/clustering/voxel_grid_based_euclidean_cluster.param.yaml]
      • radar_lanelet_filtering_range_param_path [default: $(var perception_config_dir)/object_recognition/detection/object_filter/radar_lanelet_filter.param.yaml]
      • object_recognition_detection_object_sorter_radar_param_path [default: $(var perception_config_dir)/object_recognition/detection/radar_filter/detected_object_sorter.param.yaml]
      • object_recognition_tracking_multi_object_tracker_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/data_association_matrix.param.yaml]
      • object_recognition_tracking_multi_object_tracker_input_channels_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/input_channels.param.yaml]
      • object_recognition_tracking_multi_object_tracker_node_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/multi_object_tracker_node.param.yaml]
      • euclidean_clustering/param_path [default: $(var perception_config_dir)/object_recognition/detection/euclidean_cluster/label_based_euclidean_cluster.param.yaml]
  • launch/object_recognition/prediction/prediction.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • use_vector_map [default: false]
      • prediction_model_type [default: map_based]
      • input/objects [default: /perception/object_recognition/tracking/objects]
      • output/objects [default: /perception/object_recognition/objects]
      • object_recognition_prediction_map_based_prediction_param_path [default: $(var perception_config_dir)/object_recognition/prediction/map_based_prediction.param.yaml]
  • launch/obstacle_segmentation/ground_segmentation_additional_lidars.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • target_container
      • lidar_name
      • base_frame [default: base_link]
      • use_intra_process [default: true]
      • input/main/pointcloud
      • input/lidar/pointcloud [default: /sensing/lidar/$(var lidar_name)/pointcloud]
      • output/pointcloud
      • output/pointcloud_info [default: $(var output/pointcloud)_info]
      • crop_box_margin_max_z [default: 2.5]
      • crop_box_margin_min_z [default: -2.5]
      • crop_box_filter_node_param_path [default: $(var perception_config_dir)/obstacle_segmentation/ground_segmentation/crop_box_filter.param.yaml]
      • scan_ground_filter_node_param_path [default: $(var perception_config_dir)/obstacle_segmentation/ground_segmentation/scan_ground_filter.param.yaml]
  • launch/obstacle_segmentation/ground_segmentation_ml.launch.xml
      • pointcloud_container_name [default: pointcloud_container]
      • use_pointcloud_container [default: true]
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • output/pointcloud [default: /perception/obstacle_segmentation/pointcloud]
      • output/pointcloud/segmentation [default: /perception/segmented/pointcloud]
      • output/pointcloud/visualization [default: /perception/segmented/pointcloud/visualization]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • model_variant [default: ptv3]
      • model_path [default: $(var data_path)/ptv3]
  • launch/obstacle_segmentation/ground_segmentation_ransac.launch.xml
      • target_container
      • base_frame [default: base_link]
      • use_intra_process [default: true]
      • input_topics
      • input/main/pointcloud
      • output/pointcloud
      • output/pointcloud_info [default: $(var output/pointcloud)_info]
      • detection_area_crop_box_filter_node_param_path
      • ransac_ground_filter_node_param_path [default: $(find-pkg-share autoware_ground_segmentation)/config/ransac_ground_filter.param.yaml]
  • launch/obstacle_segmentation/ground_segmentation_rule_based.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • target_container
      • base_frame [default: base_link]
      • use_intra_process [default: true]
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • output/pointcloud [default: /perception/obstacle_segmentation/pointcloud]
      • crop_box_margin_max_z [default: 2.5]
      • crop_box_margin_min_z [default: -2.5]
      • crop_box_filter_node_param_path [default: $(var perception_config_dir)/obstacle_segmentation/ground_segmentation/crop_box_filter.param.yaml]
      • scan_ground_filter_node_param_path [default: $(var perception_config_dir)/obstacle_segmentation/ground_segmentation/scan_ground_filter.param.yaml]
  • launch/obstacle_segmentation/ground_segmentation_rule_based_cuda.launch.xml
      • target_container
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • output/pointcloud [default: /perception/obstacle_segmentation/pointcloud]
      • output/ground_pointcloud [default: /perception/obstacle_segmentation/ground_pointcloud]
      • cuda_scan_ground_segmentation_filter_node_param_path [default: $(find-pkg-share autoware_ground_segmentation_cuda)/config/cuda_scan_ground_segmentation_filter.param.yaml]
  • launch/occupancy_grid_map/probabilistic_occupancy_grid_map.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • input/obstacle_pointcloud [default: concatenated/pointcloud]
      • input/raw_pointcloud [default: no_ground/oneshot/pointcloud]
      • output [default: /perception/occupancy_grid_map/map]
      • use_intra_process [default: false]
      • use_multithread [default: false]
      • pointcloud_container_name [default: pointcloud_container]
      • occupancy_grid_map_method
      • occupancy_grid_map_param_path [default: $(var perception_config_dir)/occupancy_grid_map/$(var occupancy_grid_map_method).param.yaml]
      • occupancy_grid_map_updater
      • occupancy_grid_map_updater_param_path [default: $(var perception_config_dir)/occupancy_grid_map/$(var occupancy_grid_map_updater)_updater.param.yaml]
      • input_obstacle_pointcloud [default: false]
      • input_obstacle_and_raw_pointcloud [default: true]
      • use_pointcloud_container [default: true]
  • launch/traffic_light_recognition/traffic_light_detector_fine.launch.xml
    • Fine-detection chain for one traffic light camera: map based detector proposes rough ROIs, the fine detector refines them, classifiers label them.
      • perception_config_pkg [default: autoware_perception_config]
      • camera_namespace
      • enable_image_decompressor [default: true]
      • input/vector_map [default: /map/vector_map]
      • input/route [default: /planning/mission_planning/route]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • map_based_detector_param_path [default: $(find-pkg-share autoware_traffic_light_map_based_detector)/config/traffic_light_map_based_detector.param.yaml]
      • fine_detector_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_fine_detector/traffic_light_fine_detector.param.yaml]
      • car_classifier_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_classifier/car_traffic_light_classifier.param.yaml]
      • pedestrian_classifier_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_classifier/pedestrian_traffic_light_classifier.param.yaml]
      • lamp_recognizer_ml_param_path [default: $(var data_path)/traffic_light_classifier/lamp_recognizer_ml.param.yaml]
      • roi_visualizer_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_visualization/traffic_light_roi_visualizer.param.yaml]
      • fine_detector_model_path [default: $(var data_path)/traffic_light_fine_detector/tlr_car_ped_yolox_s_batch_6.onnx]
      • fine_detector_label_path [default: $(var data_path)/traffic_light_fine_detector/tlr_labels.txt]
      • car_classifier_model_path [default: $(var data_path)/traffic_light_classifier/traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • car_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels.txt]
      • pedestrian_classifier_model_path [default: $(var data_path)/traffic_light_classifier/ped_traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • pedestrian_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels_ped.txt]
      • car_classifier_type [default: 1]
      • pedestrian_classifier_type [default: 1]
  • launch/traffic_light_recognition/traffic_light_detector_rough.launch.xml
    • Rough-detection chain for one traffic light camera: map based detector ROIs feed the classifiers directly, without a high accuracy detector.
      • perception_config_pkg [default: autoware_perception_config]
      • camera_namespace
      • enable_image_decompressor [default: true]
      • input/vector_map [default: /map/vector_map]
      • input/route [default: /planning/mission_planning/route]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • map_based_detector_param_path [default: $(find-pkg-share autoware_traffic_light_map_based_detector)/config/traffic_light_map_based_detector.param.yaml]
      • car_classifier_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_classifier/car_traffic_light_classifier.param.yaml]
      • pedestrian_classifier_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_classifier/pedestrian_traffic_light_classifier.param.yaml]
      • lamp_recognizer_ml_param_path [default: $(var data_path)/traffic_light_classifier/lamp_recognizer_ml.param.yaml]
      • roi_visualizer_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_visualization/traffic_light_roi_visualizer.param.yaml]
      • car_classifier_model_path [default: $(var data_path)/traffic_light_classifier/traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • car_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels.txt]
      • pedestrian_classifier_model_path [default: $(var data_path)/traffic_light_classifier/ped_traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • pedestrian_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels_ped.txt]
      • car_classifier_type [default: 1]
      • pedestrian_classifier_type [default: 1]
  • launch/traffic_light_recognition/traffic_light_detector_whole_image.launch.xml
    • Whole-image detection chain for one traffic light camera: yolox detects over the full image, the selector matches detections to map based ROIs, classifiers label them.
      • perception_config_pkg [default: autoware_perception_config]
      • camera_namespace
      • enable_image_decompressor [default: true]
      • input/vector_map [default: /map/vector_map]
      • input/route [default: /planning/mission_planning/route]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • map_based_detector_param_path [default: $(find-pkg-share autoware_traffic_light_map_based_detector)/config/traffic_light_map_based_detector.param.yaml]
      • whole_image_detector_param_path [default: $(var perception_config_dir)/traffic_light_recognition/tensorrt_yolox/yolox_traffic_light_detector.param.yaml]
      • car_classifier_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_classifier/car_traffic_light_classifier.param.yaml]
      • pedestrian_classifier_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_classifier/pedestrian_traffic_light_classifier.param.yaml]
      • lamp_recognizer_ml_param_path [default: $(var data_path)/traffic_light_classifier/lamp_recognizer_ml.param.yaml]
      • roi_visualizer_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_visualization/traffic_light_roi_visualizer.param.yaml]
      • whole_image_detector_model_path [default: $(var data_path)/tensorrt_yolox/yolox_s_car_ped_tl_detector_960_960_batch_1.onnx]
      • whole_image_detector_label_path [default: $(var data_path)/tensorrt_yolox/car_ped_tl_detector_labels.txt]
      • car_classifier_model_path [default: $(var data_path)/traffic_light_classifier/traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • car_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels.txt]
      • pedestrian_classifier_model_path [default: $(var data_path)/traffic_light_classifier/ped_traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • pedestrian_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels_ped.txt]
      • car_classifier_type [default: 1]
      • pedestrian_classifier_type [default: 1]
  • launch/traffic_light_recognition/traffic_light_fusion.launch.xml
    • Post-process shared by all traffic light cameras: multi camera fusion, V2X arbitration, pedestrian signal estimation and map visualization.
      • perception_config_pkg [default: autoware_perception_config]
      • camera_namespaces [default: [camera6, camera7]]
      • input/vector_map [default: /map/vector_map]
      • external/traffic_signals [default: /perception/traffic_light_recognition/external/traffic_signals]
      • internal/traffic_signals [default: /perception/traffic_light_recognition/internal/traffic_signals]
      • judged/traffic_signals [default: /perception/traffic_light_recognition/judged/traffic_signals]
      • output/traffic_signals [default: /perception/traffic_light_recognition/traffic_signals]
      • multi_camera_fusion_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_multi_camera_fusion/traffic_light_multi_camera_fusion.param.yaml]
      • arbiter_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_arbiter/traffic_light_arbiter.param.yaml]
      • crosswalk_estimator_param_path [default: $(var perception_config_dir)/traffic_light_recognition/crosswalk_traffic_light_estimator/crosswalk_traffic_light_estimator.param.yaml]
  • launch/traffic_light_recognition/traffic_light_occlusion_predictor.launch.xml
    • Occlusion check for one traffic light camera: merges the car and pedestrian signals into the per-camera output, masking signals hidden from the camera.
      • perception_config_pkg [default: autoware_perception_config]
      • camera_namespace
      • input/cloud [default: /sensing/lidar/top/pointcloud_raw_ex]
      • input/vector_map [default: /map/vector_map]
      • occlusion_predictor_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_occlusion_predictor/traffic_light_occlusion_predictor.param.yaml]
  • launch/traffic_light_recognition/traffic_light_recognition_fine.launch.xml
    • Traffic light recognition, fine detection mode: per-camera fine detection with occlusion check, then shared fusion. Two cameras as the representative set; add or remove per-camera blocks to match the vehicle.
      • perception_config_pkg [default: autoware_perception_config]
      • input/vector_map [default: /map/vector_map]
      • input/route [default: /planning/mission_planning/route]
      • input/pointcloud [default: /sensing/lidar/top/pointcloud_raw_ex]
      • external/traffic_signals [default: /perception/traffic_light_recognition/external/traffic_signals]
      • output/traffic_signals [default: /perception/traffic_light_recognition/traffic_signals]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • fine_detector_model_path [default: $(var data_path)/traffic_light_fine_detector/tlr_car_ped_yolox_s_batch_6.onnx]
      • fine_detector_label_path [default: $(var data_path)/traffic_light_fine_detector/tlr_labels.txt]
      • lamp_recognizer_ml_param_path [default: $(var data_path)/traffic_light_classifier/lamp_recognizer_ml.param.yaml]
      • car_classifier_model_path [default: $(var data_path)/traffic_light_classifier/traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • car_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels.txt]
      • pedestrian_classifier_model_path [default: $(var data_path)/traffic_light_classifier/ped_traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • pedestrian_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels_ped.txt]
      • car_classifier_type [default: 1]
      • pedestrian_classifier_type [default: 1]
      • base_map_based_detector_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_map_based_detector/base_traffic_light_map_based_detector.param.yaml]
      • extra_map_based_detector_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_map_based_detector/extra_traffic_light_map_based_detector.param.yaml]
  • launch/traffic_light_recognition/traffic_light_recognition_merge.launch.xml
    • Traffic light recognition, merge mode: detection and classification run on other ECUs; this side runs the per-camera occlusion check and the shared fusion. Two cameras as the representative set; add or remove per-camera blocks to match the vehicle.
      • input/vector_map [default: /map/vector_map]
      • input/pointcloud [default: /sensing/lidar/top/pointcloud_raw_ex]
      • external/traffic_signals [default: /perception/traffic_light_recognition/external/traffic_signals]
      • output/traffic_signals [default: /perception/traffic_light_recognition/traffic_signals]
  • launch/traffic_light_recognition/traffic_light_recognition_rough.launch.xml
    • Traffic light recognition, rough detection mode: per-camera map based detection with occlusion check, then shared fusion. One camera as the representative set; add or remove per-camera blocks to match the vehicle.
      • perception_config_pkg [default: autoware_perception_config]
      • input/vector_map [default: /map/vector_map]
      • input/route [default: /planning/mission_planning/route]
      • input/pointcloud [default: /sensing/lidar/top/pointcloud_raw_ex]
      • external/traffic_signals [default: /perception/traffic_light_recognition/external/traffic_signals]
      • output/traffic_signals [default: /perception/traffic_light_recognition/traffic_signals]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • lamp_recognizer_ml_param_path [default: $(var data_path)/traffic_light_classifier/lamp_recognizer_ml.param.yaml]
      • car_classifier_model_path [default: $(var data_path)/traffic_light_classifier/traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • car_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels.txt]
      • pedestrian_classifier_model_path [default: $(var data_path)/traffic_light_classifier/ped_traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • pedestrian_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels_ped.txt]
      • car_classifier_type [default: 1]
      • pedestrian_classifier_type [default: 1]
      • base_map_based_detector_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_map_based_detector/base_traffic_light_map_based_detector.param.yaml]
  • launch/traffic_light_recognition/traffic_light_recognition_whole_image.launch.xml
    • Traffic light recognition, whole-image detection mode: per-camera yolox detection, then shared fusion. Two cameras as the representative set; add or remove per-camera blocks to match the vehicle.
      • perception_config_pkg [default: autoware_perception_config]
      • input/vector_map [default: /map/vector_map]
      • input/route [default: /planning/mission_planning/route]
      • external/traffic_signals [default: /perception/traffic_light_recognition/external/traffic_signals]
      • output/traffic_signals [default: /perception/traffic_light_recognition/traffic_signals]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • whole_image_detector_model_path [default: $(var data_path)/tensorrt_yolox/yolox_s_car_ped_tl_detector_960_960_batch_1.onnx]
      • whole_image_detector_label_path [default: $(var data_path)/tensorrt_yolox/car_ped_tl_detector_labels.txt]
      • lamp_recognizer_ml_param_path [default: $(var data_path)/traffic_light_classifier/lamp_recognizer_ml.param.yaml]
      • car_classifier_model_path [default: $(var data_path)/traffic_light_classifier/traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • car_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels.txt]
      • pedestrian_classifier_model_path [default: $(var data_path)/traffic_light_classifier/ped_traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • pedestrian_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels_ped.txt]
      • car_classifier_type [default: 1]
      • pedestrian_classifier_type [default: 1]
      • base_map_based_detector_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_map_based_detector/base_traffic_light_map_based_detector.param.yaml]
      • extra_map_based_detector_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_map_based_detector/extra_traffic_light_map_based_detector.param.yaml]

Messages

No message files found.

Services

No service files found

Plugins

No plugins found.

Recent questions tagged autoware_perception_launch at Robotics Stack Exchange

Package symbol

autoware_perception_launch package from autoware_launch repo

autoware_launch autoware_sample_designs autoware_control_config autoware_control_launch autoware_localization_config autoware_localization_launch autoware_map_config autoware_map_launch autoware_perception_config autoware_perception_launch autoware_planning_config autoware_planning_launch autoware_sensing_launch autoware_sensor_driver_launch awsim_labs_sensor_kit_description awsim_labs_sensor_kit_launch awsim_sensor_kit_description awsim_sensor_kit_launch carla_sensor_kit_description carla_sensor_kit_launch sample_sensor_kit_description sample_sensor_kit_launch single_lidar_sensor_kit_description single_lidar_sensor_kit_launch tier4_autoware_api_launch tier4_simulator_launch tier4_system_launch tier4_vehicle_launch awsim_labs_vehicle_description awsim_labs_vehicle_launch sample_vehicle_description sample_vehicle_launch

ROS Distro
github

API Docs
Browse Code

Package Summary

Version 0.53.0
License Apache License 2.0
Build type AMENT_CMAKE
Use RECOMMENDED

Repository Summary

Description
Checkout URI https://github.com/autowarefoundation/autoware_launch.git
VCS Type git
VCS Version main
Last Updated 2026-10-26
Dev Status UNKNOWN
Released UNRELEASED
Contributing Help Wanted (-)
Good First Issues (-)
Pull Requests to Review (-)

Package Description

The autoware_perception_launch package

Maintainers

  • Yukihiro Saito
  • Dai Nguyen
  • Taekjin Lee
  • Yoshi Ri
  • Masato Saeki

Authors

No additional authors.
autoware_universe_launch/autoware_perception_launch/README.md

autoware_perception_launch

Shared perception launch library. The system component launchers live in the product layer (autoware_launch/launch/components/component_perception.launch.xml for the lidar pipeline and component_traffic_light.launch.xml for traffic light recognition); this package hosts the perception feature launchers they combine:

launch/
├── common/                        # shared preprocess feature launchers (pointcloud downsample)
├── object_recognition/
│   ├── object_recognition_<mode>.launch.xml  # detection + tracking + prediction per sensor mode
│   ├── detection/
│   │   ├── detector/              # detectors
│   │   ├── filter/                # filters
│   │   └── merger/                # detection-side serial merger (used by the camera_lidar_radar_serial launcher)
│   └── prediction/
├── obstacle_segmentation/
├── occupancy_grid_map/
└── traffic_light_recognition/
design/module/                     # Autoware System Designer modules; same subtree as launch/
├── object_recognition/
├── obstacle_segmentation/
├── occupancy_grid_map/
└── traffic_light_recognition/

Launch files

  • object_recognition/object_recognition_<mode>.launch.xml — one feature launcher per sensor mode (camera, lidar, lidar_ptv3, lidar_radar, camera_lidar_radar, camera_lidar_radar_serial); hosts the detection stage inline together with the tracker and prediction, and pushes their module namespaces. Camera and radar input topics default to their absolute sensing paths (/sensing/camera/cameraN/..., /sensing/radar/detected_objects), so callers override only what differs. Most modes wire the multi-channel tracker to the detector outputs; camera_lidar_radar_serial instead merges the detectors in the detection stage (optionally regenerating detections through detection-by-tracker), tracks the single merged detected_objects channel, and joins radar tracked objects at the decorative tracker merger. camera runs camera-only detectors (BEV and camera VRU).
  • The clustering backend is file selection, not a flag: lidar runs rule-based euclidean clustering off the obstacle segmentation cloud, while lidar_ptv3 and lidar_radar cluster the PTv3 semantic-segmentation cloud and drop the pointcloud map filter and the object validator that only the rule-based pipeline uses. Both feed their cluster objects to the tracker on the lidar_clustering channel.
  • object_recognition/detection/{detector,filter}/ — single-purpose launchers wired together by the object-recognition launchers. merger/camera_lidar_merger.launch.xml chains the detection-side object mergers for the camera_lidar_radar_serial launcher.
  • obstacle_segmentation/ground_segmentation_<backend>.launch.xml — ground removal backends (rule_based, rule_based_cuda, ml); backend selection is file selection. ground_segmentation_{additional_lidars,ransac}.launch.xml fold extra lidars into the obstacle stream for kits that carry them.
  • occupancy_grid_map/, common/, traffic_light_recognition/ — feature launchers included directly by the system components.

Namespace and container addressing

Callers own the namespaces: the system component pushes perception and object_recognition; object_recognition_<mode>.launch.xml pushes detection/tracking/prediction. The object-recognition launchers address their internal topics through the fixed /perception/object_recognition namespace. Composable nodes load into the shared pointcloud container, whose name is exposed as pointcloud_container_name and aliased to the node/pointcloud_container argument of the detector launchers.

Config

Parameter files are resolved from a config/ tree addressed through the perception_config_pkg argument, which defaults to autoware_perception_config. A product can relocate the whole tree by setting perception_config_pkg once at its entry point — the launch configuration propagates through every include — but it must then ship the same config/ layout. Products that differ in only a few files instead override the corresponding *_param_path arguments, which every feature launcher declares with a default:

<include file="$(find-pkg-share autoware_perception_launch)/launch/object_recognition/object_recognition_lidar.launch.xml">
  <arg name="perception_config_pkg" value="my_perception_config"/>
  <arg name="FOO_NODE_param_path" value="..."/>
</include>

component_perception.launch.xml, component_traffic_light.launch.xml and planning_simulator/perception_simulation.launch.xml each declare perception_config_pkg, so a product switches the tree at the component it launches.

The launch/ tree is the interface downstream distributions adopt as-is; only package.xml, CMakeLists.txt, README.md, and changelogs stay package-local. Keep changes to this tree folder-merge compatible.

Package Dependencies

Please see <exec_depend> in package.xml.

CHANGELOG

Changelog for package autoware_perception_launch

0.53.0 (2026-10-13)

  • Merge remote-tracking branch 'origin/main' into tmp/bot/bump_version_base

  • feat(autoware_launch): run crop_box_filter standalone when agnocast is enabled (#1990)

    * feat(autoware_launch): run crop_box_filter standalone when agnocast is enabled autoware_universe#13384 moves CropBoxFilterComponent to agnocast_wrapper::Node on an agnocast-only executor, which the component containers do not set up. Start each of the eight crop_box_filter nodes as a standalone process when ENABLE_AGNOCAST=1 and keep loading them into their container otherwise. The node names, namespaces, remaps and parameters are the same in both branches.

    * refactor(autoware_launch): launch crop_box_filter through the autoware_node action ---------

  • feat(autoware_perception_launch): run scan_ground_filter standalone when agnocast is enabled (#1991)

    * feat(autoware_perception_launch): run scan_ground_filter standalone when agnocast is enabled autoware_universe#13385 moves ScanGroundFilterComponent to agnocast_wrapper::Node on an agnocast-only executor, which the component containers do not set up. Start each of the three scan_ground_filter nodes as a standalone process when ENABLE_AGNOCAST=1 and keep loading them into their container otherwise. The node names, namespaces, remaps and parameters are the same in both branches.

    * refactor(autoware_perception_launch): launch scan_ground_filter through the autoware_node action ---------

  • feat(autoware_launch): run voxel_grid_downsample_filter standalone when agnocast is enabled (#1992)

    • feat(autoware_launch): run voxel_grid_downsample_filter standalone when agnocast is enabled
    • style(pre-commit): autofix

    * refactor(autoware_launch): launch voxel_grid_downsample_filter through the autoware_node action ---------Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>

  • feat(autoware_sample_designs): compose the AD API, system and vehicle components from design modules (#1976)

    * fix(control, planning): connect the operation mode state and velocity limit inputs Wires collision_detector's operation_mode_state and remaining_distance_time_calculator's planning_velocity through the module designs; their node designs now declare the fixed topic names as remap targets.

    * fix(planning): connect the internal velocity limit path to the selector Forwards motion_velocity_planner's velocity_limit and clear_velocity_limit up to ScenarioPlanning and feeds them into external_velocity_limit_selector, whose internal inputs were unconnected.

    * fix(control, planning): connect the remaining module inputs to their sources Wires the operation mode transition manager's gate and selector inputs, the stop mode operator's vehicle status, the manual lane change handler's route availability and preferred primitive service, the parking costmap, and the route marker. Exposes the vehicle control mode service, the preferred lane service and the system hazard lights command as component interfaces.

    • fix(planning): follow the hazard lights selector input rename
    • fix(planning): add steering_status subscriber to various planning modules
    • feat(vehicle): add vehicle wrapper and launch file for vehicle interface
    • style(pre-commit): autofix
    • fix(planning): add planning factors publishers and connections for various modules
    • refactor(autoware_sample_designs): compose the AD API from design modules
    • Refactor ADAPI module structure and enhance planning factors
    • Updated AdapiRvizHelpers.module.yaml to include helper prefixes for instances and connections.
    • Modified AutowareSample.system.yaml to reorganize ADAPI components, introducing default_adapi for better structure.
    • Expanded planning factors in Planning.module.yaml, LaneDriving.module.yaml, ScenarioPlanning.module.yaml, and associated modules to include additional factors for improved behavior and motion planning.
    • Enhanced connections in behavior planning and motion planning modules to accommodate new planning factors.
    • Ensured consistency across scenario planning modules by aligning

File truncated at 100 lines see the full file

Package Dependencies

Deps Name
ament_cmake_auto
autoware_cmake
autoware_agnocast_wrapper
autoware_bevfusion
autoware_bytetrack
autoware_compare_map_segmentation
autoware_crosswalk_traffic_light_estimator
autoware_cuda_pointcloud_preprocessor
autoware_detected_object_feature_remover
autoware_detected_object_validation
autoware_detection_by_tracker
autoware_elevation_map_loader
autoware_euclidean_cluster
autoware_ground_segmentation
autoware_ground_segmentation_cuda
autoware_image_object_locator
autoware_image_projection_based_fusion
autoware_image_transport_decompressor
autoware_lidar_apollo_instance_segmentation
autoware_lidar_centerpoint
autoware_lidar_transfusion
autoware_map_based_prediction
autoware_multi_object_tracker
autoware_object_merger
autoware_object_sorter
autoware_occupancy_grid_map_outlier_filter
autoware_perception_config
autoware_pointcloud_preprocessor
autoware_probabilistic_occupancy_grid_map
autoware_ptv3
autoware_raindrop_cluster_filter
autoware_shape_estimation
autoware_simpl_prediction
autoware_simple_object_merger
autoware_tensorrt_bevdet
autoware_tensorrt_yolox
autoware_tracking_object_merger
autoware_traffic_light_arbiter
autoware_traffic_light_category_merger
autoware_traffic_light_classifier
autoware_traffic_light_fine_detector
autoware_traffic_light_map_based_detector
autoware_traffic_light_multi_camera_fusion
autoware_traffic_light_occlusion_predictor
autoware_traffic_light_selector
autoware_traffic_light_visualization
rclcpp_components
topic_tools
ament_lint_auto
autoware_lint_common

System Dependencies

No direct system dependencies.

Dependant Packages

Name Deps
autoware_launch
reaction_analyzer

Launch files

  • launch/common/pointcloud_downsample.launch.xml
      • pointcloud_container_name
      • cuda_pointcloud_preprocessing [default: false]
      • input/pointcloud
      • output/pointcloud
      • common_downsample_voxel_size_x [default: 0.05]
      • common_downsample_voxel_size_y [default: 0.05]
      • common_downsample_voxel_size_z [default: 0.05]
  • launch/object_recognition/detection/detector/camera_2d_detector.launch.xml
      • image_raw0 [default: /sensing/camera/camera0/image_raw]
      • image_raw1 [default: /sensing/camera/camera1/image_raw]
      • image_raw2 [default: /sensing/camera/camera2/image_raw]
      • image_raw3 [default: /sensing/camera/camera3/image_raw]
      • image_raw4 [default: /sensing/camera/camera4/image_raw]
      • image_raw5 [default: /sensing/camera/camera5/image_raw]
      • image_raw6 [default: /sensing/camera/camera6/image_raw]
      • image_raw7 [default: /sensing/camera/camera7/image_raw]
      • image_raw8 [default: /sensing/camera/camera8/image_raw]
      • image_raw9 [default: /sensing/camera/camera9/image_raw]
      • image_number [default: 1]
      • camera_index [default: 0]
      • use_bytetrack [default: true]
      • enable_visualizer [default: false]
      • camera_2d_detector/model_path
      • camera_2d_detector/label_path
      • camera_2d_detector/color_map_path
      • camera_2d_detector/roi_remap_path
      • camera_2d_detector/roi_to_semseg_remap_path
      • tensorrt_yolox_ns [default: ]
  • launch/object_recognition/detection/detector/camera_bev_detector.launch.xml
      • input/camera0/image
      • input/camera0/info
      • input/camera1/image
      • input/camera1/info
      • input/camera2/image
      • input/camera2/info
      • input/camera3/image
      • input/camera3/info
      • input/camera4/image
      • input/camera4/info
      • input/camera5/image
      • input/camera5/info
      • output/objects
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • bevdet_model_name [default: bevdet_one_lt_d]
      • bevdet_model_path [default: $(var data_path)/tensorrt_bevdet]
  • launch/object_recognition/detection/detector/camera_lidar_detector.launch.xml
      • ns
      • lidar_detection_model_type
      • lidar_detection_model_name
      • use_low_intensity_cluster_filter
      • use_image_segmentation_based_filter
      • number_of_cameras
      • input/camera0/image
      • input/camera0/info
      • input/camera0/rois
      • input/camera1/image
      • input/camera1/info
      • input/camera1/rois
      • input/camera2/image
      • input/camera2/info
      • input/camera2/rois
      • input/camera3/image
      • input/camera3/info
      • input/camera3/rois
      • input/camera4/image
      • input/camera4/info
      • input/camera4/rois
      • input/camera5/image
      • input/camera5/info
      • input/camera5/rois
      • input/camera6/image
      • input/camera6/info
      • input/camera6/rois
      • input/camera7/image
      • input/camera7/info
      • input/camera7/rois
      • input/camera8/image
      • input/camera8/info
      • input/camera8/rois
      • input/concatenation_info
      • segmentation_pointcloud_fusion_camera_ids
      • image_topic_name
      • sync_param_path
      • voxel_grid_based_euclidean_param_path
      • lidar_model_param_path
      • pointpainting_model_path [default: $(var data_path)/image_projection_based_fusion]
      • pointpainting_fusion_common_param_path
      • roi_cluster_fusion_param_path
      • node/pointcloud_container
      • input/pointcloud
      • input/pointcloud_map/pointcloud
      • input/obstacle_segmentation/pointcloud
      • output/ml_detector/objects
      • output/rule_detector/objects
      • camera_2d_detector/model_path
      • camera_2d_detector/label_path
      • camera_2d_detector/color_map_path
      • camera_2d_detector/roi_remap_path
      • camera_2d_detector/roi_to_semseg_remap_path
      • enable_2d_detection [default: false]
  • launch/object_recognition/detection/detector/camera_lidar_irregular_object_detector.launch.xml
      • pipeline_ns
      • node/pointcloud_container
      • output/objects [default: objects]
      • input/concatenation_info
      • input/pointcloud
      • irregular_object_crop_box_filter_node_param_path
      • irregular_object_scan_ground_filter_node_param_path
      • irregular_object_roi_pointcloud_fusion_node_param_path
      • irregular_object_fusion_camera_param_path
      • sync_param_path
  • launch/object_recognition/detection/detector/camera_vru_detector.launch.xml
      • input/camera0/info [default: /sensing/camera/camera0/camera_info]
      • input/camera0/rois [default: /perception/object_recognition/detection/rois0]
      • input/camera1/info [default: /sensing/camera/camera1/camera_info]
      • input/camera1/rois [default: /perception/object_recognition/detection/rois1]
      • input/camera2/info [default: /sensing/camera/camera2/camera_info]
      • input/camera2/rois [default: /perception/object_recognition/detection/rois2]
      • input/camera3/info [default: /sensing/camera/camera3/camera_info]
      • input/camera3/rois [default: /perception/object_recognition/detection/rois3]
      • input/camera4/info [default: /sensing/camera/camera4/camera_info]
      • input/camera4/rois [default: /perception/object_recognition/detection/rois4]
      • input/camera5/info [default: /sensing/camera/camera5/camera_info]
      • input/camera5/rois [default: /perception/object_recognition/detection/rois5]
      • input/camera6/info [default: /sensing/camera/camera6/camera_info]
      • input/camera6/rois [default: /perception/object_recognition/detection/rois6]
      • input/camera7/info [default: /sensing/camera/camera7/camera_info]
      • input/camera7/rois [default: /perception/object_recognition/detection/rois7]
      • output/objects [default: /perception/object_recognition/detection/camera_vru/objects]
      • bbox_object_locator_param_path
      • rois_ids [default: [0, 1]]
  • launch/object_recognition/detection/detector/irregular_object_detector.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • target_container
      • base_frame [default: base_link]
      • use_intra_process [default: true]
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • input/concatenation_info [default: /sensing/lidar/concatenated/pointcloud_info]
      • output/clusters
      • crop_box_filter_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/crop_box_filter.param.yaml]
      • scan_ground_filter_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/scan_ground_filter.param.yaml]
      • roi_pointcloud_fusion_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/roi_pointcloud_fusion.param.yaml]
      • fusion_camera_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/fusion_cameras.param.yaml]
      • sync_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/fusion_common.param.yaml]
  • launch/object_recognition/detection/detector/lidar_dnn_detector.launch.xml
      • lidar_detection_model_type
      • lidar_detection_model_name
      • lidar_model_param_path
      • bevfusion_model_path [default: $(var data_path)/bevfusion]
      • centerpoint_model_path [default: $(var data_path)/lidar_centerpoint]
      • centerpoint_common_param_path [default: $(find-pkg-share autoware_lidar_centerpoint)/config/centerpoint_common.param.yaml]
      • transfusion_model_path [default: $(var data_path)/lidar_transfusion]
      • use_short_range_detection [default: false]
      • lidar_short_range_detection_model_type
      • node/pointcloud_container
      • input/pointcloud
      • output/objects
      • output/short_range_objects
  • launch/object_recognition/detection/detector/lidar_rule_detector.launch.xml
      • ns
      • node/pointcloud_container
      • input/pointcloud_map/pointcloud
      • input/obstacle_segmentation/pointcloud
      • output/cluster_objects
      • output/objects
      • voxel_grid_based_euclidean_param_path
      • use_low_height_cropbox [default: false]
  • launch/object_recognition/detection/detector/lidar_semantic_segmentation_detector.launch.xml
      • euclidean_clustering/param_path
      • input/pointcloud
      • output/objects
  • launch/object_recognition/detection/detector/tracker_based_detector.launch.xml
      • detection_by_tracker_param_path
      • input/clusters
      • input/tracked_objects
      • output/objects
  • launch/object_recognition/detection/filter/object_filter.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • objects_filter_method [default: lanelet_filter]
      • object_recognition_detection_object_lanelet_filter_param_path [default: $(var perception_config_dir)/object_recognition/detection/object_filter/object_lanelet_filter.param.yaml]
      • object_recognition_detection_object_position_filter_param_path [default: $(var perception_config_dir)/object_recognition/detection/object_filter/object_position_filter.param.yaml]
      • input/objects
      • output/objects
  • launch/object_recognition/detection/filter/object_validator.launch.xml
      • objects_validation_method
      • object_recognition_detection_obstacle_pointcloud_based_validator_param_path
      • input/obstacle_pointcloud
      • input/objects
      • output/objects
  • launch/object_recognition/detection/filter/pointcloud_downsample_filter.launch.xml
      • target_container
      • use_intra_process [default: true]
      • input/pointcloud
      • output/pointcloud
      • downsample_voxel_size [default: 0.1]
  • launch/object_recognition/detection/filter/pointcloud_map_filter.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • target_container
      • use_intra_process [default: true]
      • input/pointcloud
      • output/pointcloud
      • output/downsampled/pointcloud [default: /perception/obstacle_segmentation/pointcloud_map_filtered/downsampled/pointcloud]
      • downsample_voxel_size [default: 0.1]
      • voxel_based_compare_map_filter_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/pointcloud_filter/voxel_based_compare_map_filter.param.yaml]
  • launch/object_recognition/detection/filter/radar_filter.launch.xml
      • object_sorter_param_path
      • radar_lanelet_filtering_range_param_path
      • input/radar
      • output/objects
  • launch/object_recognition/detection/merger/camera_lidar_merger.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • object_recognition_detection_object_merger_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/detection/object_merger/data_association_matrix.param.yaml]
      • object_recognition_detection_object_merger_distance_threshold_list_path [default: $(var perception_config_dir)/object_recognition/detection/object_merger/overlapped_judge.param.yaml]
      • ml_camera_lidar_object_association_merger_param_path [default: $(var perception_config_dir)/object_recognition/detection/object_merger/ml_camera_lidar_object_association_merger.param.yaml]
      • roi_detected_object_fusion_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/roi_detected_object_fusion.param.yaml]
      • sync_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/fusion_common.param.yaml]
      • lidar_detection_model_type
      • use_detection_by_tracker
      • use_irregular_object_detector
      • use_roi_detected_object_validator
      • euclidean_clustering/shape_policy
      • use_object_filter
      • objects_filter_method
      • number_of_cameras
      • input/camera0/image
      • input/camera0/info
      • input/camera0/rois
      • input/camera1/image
      • input/camera1/info
      • input/camera1/rois
      • input/camera2/image
      • input/camera2/info
      • input/camera2/rois
      • input/camera3/image
      • input/camera3/info
      • input/camera3/rois
      • input/camera4/image
      • input/camera4/info
      • input/camera4/rois
      • input/camera5/image
      • input/camera5/info
      • input/camera5/rois
      • input/camera6/image
      • input/camera6/info
      • input/camera6/rois
      • input/camera7/image
      • input/camera7/info
      • input/camera7/rois
      • input/camera8/image
      • input/camera8/info
      • input/camera8/rois
      • input/concatenation_info [default: /sensing/lidar/concatenated/pointcloud_info]
      • input/lidar_ml/objects
      • input/lidar_cluster/objects
      • input/irregular_object/objects [default: ]
      • input/detection_by_tracker/objects
      • output/objects [default: objects]
      • alpha_merger_priority_mode [default: 0]
  • launch/object_recognition/object_recognition_camera.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • use_camera_vru_detector [default: false]
      • camera_vru_detector_rois_ids [default: [0]]
      • use_camera_bev_detector [default: true]
      • use_vector_map [default: true]
      • tracker_publish_merged_objects [default: false]
      • input/camera0/image [default: /sensing/camera/camera0/image_rect_color]
      • input/camera0/info [default: /sensing/camera/camera0/camera_info]
      • input/camera0/rois [default: /perception/object_recognition/detection/rois0]
      • input/camera1/image [default: /sensing/camera/camera1/image_rect_color]
      • input/camera1/info [default: /sensing/camera/camera1/camera_info]
      • input/camera1/rois [default: /perception/object_recognition/detection/rois1]
      • input/camera2/image [default: /sensing/camera/camera2/image_rect_color]
      • input/camera2/info [default: /sensing/camera/camera2/camera_info]
      • input/camera2/rois [default: /perception/object_recognition/detection/rois2]
      • input/camera3/image [default: /sensing/camera/camera3/image_rect_color]
      • input/camera3/info [default: /sensing/camera/camera3/camera_info]
      • input/camera3/rois [default: /perception/object_recognition/detection/rois3]
      • input/camera4/image [default: /sensing/camera/camera4/image_rect_color]
      • input/camera4/info [default: /sensing/camera/camera4/camera_info]
      • input/camera4/rois [default: /perception/object_recognition/detection/rois4]
      • input/camera5/image [default: /sensing/camera/camera5/image_rect_color]
      • input/camera5/info [default: /sensing/camera/camera5/camera_info]
      • input/camera5/rois [default: /perception/object_recognition/detection/rois5]
      • input/camera6/image [default: /sensing/camera/camera6/image_rect_color]
      • input/camera6/info [default: /sensing/camera/camera6/camera_info]
      • input/camera6/rois [default: /perception/object_recognition/detection/rois6]
      • input/camera7/image [default: /sensing/camera/camera7/image_rect_color]
      • input/camera7/info [default: /sensing/camera/camera7/camera_info]
      • input/camera7/rois [default: /perception/object_recognition/detection/rois7]
      • output/tracker_merged_objects [default: /perception/object_recognition/detection/objects]
      • output/tracked_objects [default: /perception/object_recognition/tracking/objects]
      • output/predicted_objects [default: /perception/object_recognition/objects]
      • near_range_camera_vru_param_path [default: $(var perception_config_dir)/object_recognition/detection/camera_vru_detection/near_range_camera_vru_detector.param.yaml]
      • object_recognition_tracking_multi_object_tracker_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/data_association_matrix.param.yaml]
      • object_recognition_tracking_multi_object_tracker_input_channels_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/input_channels.param.yaml]
      • object_recognition_tracking_multi_object_tracker_node_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/multi_object_tracker_node.param.yaml]
  • launch/object_recognition/object_recognition_camera_lidar_radar.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • lidar_detection_model [default: centerpoint]
      • use_short_range_detection [default: false]
      • lidar_short_range_detection_model_type [default: centerpoint_short_range]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • bevfusion_model_path [default: $(var data_path)/bevfusion]
      • pointpainting_model_path [default: $(var data_path)/image_projection_based_fusion]
      • camera_2d_detector/model_path [default: $(var data_path)/tensorrt_yolox/yolox-sPlus-opt-pseudoV2-T4-960x960-T4-seg16cls.onnx]
      • camera_2d_detector/label_path [default: $(var data_path)/tensorrt_yolox/label.txt]
      • camera_2d_detector/color_map_path [default: $(var data_path)/tensorrt_yolox/semseg_color_map.csv]
      • camera_2d_detector/roi_remap_path [default: $(var perception_config_dir)/traffic_light_recognition/tensorrt_yolox/roi_label_remap.csv]
      • camera_2d_detector/roi_to_semseg_remap_path [default: $(var perception_config_dir)/traffic_light_recognition/tensorrt_yolox/roi_to_semseg_label_remap.csv]
      • use_semantic_segmentation_ptv3 [default: false]
      • use_object_filter [default: true]
      • use_pointcloud_map [default: true]
      • use_validator [default: false]
      • objects_validation_method [default: obstacle_pointcloud]
      • use_low_height_cropbox [default: false]
      • use_low_intensity_cluster_filter [default: true]
      • use_image_segmentation_based_filter [default: false]
      • use_irregular_object_detector [default: true]
      • use_camera_vru_detector [default: false]
      • camera_vru_detector_rois_ids [default: [0]]
      • use_camera_only_detection [default: false]
      • use_vector_map [default: true]
      • tracker_publish_merged_objects [default: false]
      • number_of_cameras [default: 6]
      • pointcloud_container_name [default: pointcloud_container]
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • input/semantic_segmentation/pointcloud [default: /perception/segmented/pointcloud]
      • input/obstacle_segmentation/pointcloud [default: /perception/obstacle_segmentation/pointcloud]
      • input/camera0/image [default: /sensing/camera/camera0/image_rect_color]
      • input/camera0/info [default: /sensing/camera/camera0/camera_info]
      • input/camera0/rois [default: /perception/object_recognition/detection/rois0]
      • input/camera1/image [default: /sensing/camera/camera1/image_rect_color]
      • input/camera1/info [default: /sensing/camera/camera1/camera_info]
      • input/camera1/rois [default: /perception/object_recognition/detection/rois1]
      • input/camera2/image [default: /sensing/camera/camera2/image_rect_color]
      • input/camera2/info [default: /sensing/camera/camera2/camera_info]
      • input/camera2/rois [default: /perception/object_recognition/detection/rois2]
      • input/camera3/image [default: /sensing/camera/camera3/image_rect_color]
      • input/camera3/info [default: /sensing/camera/camera3/camera_info]
      • input/camera3/rois [default: /perception/object_recognition/detection/rois3]
      • input/camera4/image [default: /sensing/camera/camera4/image_rect_color]
      • input/camera4/info [default: /sensing/camera/camera4/camera_info]
      • input/camera4/rois [default: /perception/object_recognition/detection/rois4]
      • input/camera5/image [default: /sensing/camera/camera5/image_rect_color]
      • input/camera5/info [default: /sensing/camera/camera5/camera_info]
      • input/camera5/rois [default: /perception/object_recognition/detection/rois5]
      • input/camera6/image [default: /sensing/camera/camera6/image_rect_color]
      • input/camera6/info [default: /sensing/camera/camera6/camera_info]
      • input/camera6/rois [default: /perception/object_recognition/detection/rois6]
      • input/camera7/image [default: /sensing/camera/camera7/image_rect_color]
      • input/camera7/info [default: /sensing/camera/camera7/camera_info]
      • input/camera7/rois [default: /perception/object_recognition/detection/rois7]
      • input/camera8/image [default: /sensing/camera/camera8/image_rect_color]
      • input/camera8/info [default: /sensing/camera/camera8/camera_info]
      • input/camera8/rois [default: /perception/object_recognition/detection/rois8]
      • input/concatenation_info [default: /sensing/lidar/concatenated/pointcloud_info]
      • input/radar [default: /sensing/radar/detected_objects]
      • input/camera_only/objects [default: /perception/object_recognition/detection/camera_only/objects]
      • input/camera_only/channel [default: camera_streampetr]
      • image_topic_name [default: image_rect_color]
      • segmentation_pointcloud_fusion_camera_ids [default: [0,2,4]]
      • output/tracker_merged_objects [default: /perception/object_recognition/detection/objects]
      • output/tracked_objects [default: /perception/object_recognition/tracking/objects]
      • output/predicted_objects [default: /perception/object_recognition/objects]
      • sync_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/fusion_common.param.yaml]
      • roi_cluster_fusion_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/roi_cluster_fusion.param.yaml]
      • pointpainting_fusion_common_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/pointpainting_common.param.yaml]
      • irregular_object_crop_box_filter_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/crop_box_filter.param.yaml]
      • irregular_object_scan_ground_filter_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/scan_ground_filter.param.yaml]
      • irregular_object_roi_pointcloud_fusion_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/roi_pointcloud_fusion.param.yaml]
      • irregular_object_fusion_camera_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/fusion_cameras.param.yaml]
      • lidar_model_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model]
      • centerpoint_common_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model/centerpoint_common.param.yaml]
      • voxel_grid_based_euclidean_param_path [default: $(var perception_config_dir)/object_recognition/detection/clustering/voxel_grid_based_euclidean_cluster.param.yaml]
      • radar_lanelet_filtering_range_param_path [default: $(var perception_config_dir)/object_recognition/detection/object_filter/radar_lanelet_filter.param.yaml]
      • object_recognition_detection_object_sorter_radar_param_path [default: $(var perception_config_dir)/object_recognition/detection/radar_filter/detected_object_sorter.param.yaml]
      • near_range_camera_vru_param_path [default: $(var perception_config_dir)/object_recognition/detection/camera_vru_detection/near_range_camera_vru_detector.param.yaml]
      • object_recognition_detection_obstacle_pointcloud_based_validator_param_path [default: $(var perception_config_dir)/object_recognition/detection/detected_object_validation/obstacle_pointcloud_based_validator.param.yaml]
      • object_recognition_tracking_multi_object_tracker_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/data_association_matrix.param.yaml]
      • object_recognition_tracking_multi_object_tracker_input_channels_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/input_channels.param.yaml]
      • object_recognition_tracking_multi_object_tracker_node_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/multi_object_tracker_node.param.yaml]
      • euclidean_clustering/param_path [default: $(var perception_config_dir)/object_recognition/detection/euclidean_cluster/label_based_euclidean_cluster.param.yaml]
  • launch/object_recognition/object_recognition_camera_lidar_radar_serial.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • lidar_detection_model [default: centerpoint]
      • use_short_range_detection [default: false]
      • lidar_short_range_detection_model_type [default: centerpoint_short_range]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • bevfusion_model_path [default: $(var data_path)/bevfusion]
      • pointpainting_model_path [default: $(var data_path)/image_projection_based_fusion]
      • camera_2d_detector/model_path [default: $(var data_path)/tensorrt_yolox/yolox-sPlus-opt-pseudoV2-T4-960x960-T4-seg16cls.onnx]
      • camera_2d_detector/label_path [default: $(var data_path)/tensorrt_yolox/label.txt]
      • camera_2d_detector/color_map_path [default: $(var data_path)/tensorrt_yolox/semseg_color_map.csv]
      • camera_2d_detector/roi_remap_path [default: $(var perception_config_dir)/traffic_light_recognition/tensorrt_yolox/roi_label_remap.csv]
      • camera_2d_detector/roi_to_semseg_remap_path [default: $(var perception_config_dir)/traffic_light_recognition/tensorrt_yolox/roi_to_semseg_label_remap.csv]
      • use_object_filter [default: true]
      • objects_filter_method [default: lanelet_filter]
      • use_pointcloud_map [default: true]
      • use_validator [default: false]
      • objects_validation_method [default: obstacle_pointcloud]
      • use_roi_detected_object_validator [default: true]
      • use_detection_by_tracker [default: true]
      • use_low_height_cropbox [default: false]
      • use_low_intensity_cluster_filter [default: true]
      • use_image_segmentation_based_filter [default: false]
      • use_irregular_object_detector [default: true]
      • ml_camera_lidar_merger_priority_mode [default: 0]
      • use_vector_map [default: true]
      • number_of_cameras [default: 6]
      • pointcloud_container_name [default: pointcloud_container]
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • input/obstacle_segmentation/pointcloud [default: /perception/obstacle_segmentation/pointcloud]
      • input/camera0/image [default: /sensing/camera/camera0/image_rect_color]
      • input/camera0/info [default: /sensing/camera/camera0/camera_info]
      • input/camera0/rois [default: /perception/object_recognition/detection/rois0]
      • input/camera1/image [default: /sensing/camera/camera1/image_rect_color]
      • input/camera1/info [default: /sensing/camera/camera1/camera_info]
      • input/camera1/rois [default: /perception/object_recognition/detection/rois1]
      • input/camera2/image [default: /sensing/camera/camera2/image_rect_color]
      • input/camera2/info [default: /sensing/camera/camera2/camera_info]
      • input/camera2/rois [default: /perception/object_recognition/detection/rois2]
      • input/camera3/image [default: /sensing/camera/camera3/image_rect_color]
      • input/camera3/info [default: /sensing/camera/camera3/camera_info]
      • input/camera3/rois [default: /perception/object_recognition/detection/rois3]
      • input/camera4/image [default: /sensing/camera/camera4/image_rect_color]
      • input/camera4/info [default: /sensing/camera/camera4/camera_info]
      • input/camera4/rois [default: /perception/object_recognition/detection/rois4]
      • input/camera5/image [default: /sensing/camera/camera5/image_rect_color]
      • input/camera5/info [default: /sensing/camera/camera5/camera_info]
      • input/camera5/rois [default: /perception/object_recognition/detection/rois5]
      • input/camera6/image [default: /sensing/camera/camera6/image_rect_color]
      • input/camera6/info [default: /sensing/camera/camera6/camera_info]
      • input/camera6/rois [default: /perception/object_recognition/detection/rois6]
      • input/camera7/image [default: /sensing/camera/camera7/image_rect_color]
      • input/camera7/info [default: /sensing/camera/camera7/camera_info]
      • input/camera7/rois [default: /perception/object_recognition/detection/rois7]
      • input/camera8/image [default: /sensing/camera/camera8/image_rect_color]
      • input/camera8/info [default: /sensing/camera/camera8/camera_info]
      • input/camera8/rois [default: /perception/object_recognition/detection/rois8]
      • input/concatenation_info [default: /sensing/lidar/concatenated/pointcloud_info]
      • input/radar/tracked_objects [default: /sensing/radar/tracked_objects]
      • image_topic_name [default: image_rect_color]
      • segmentation_pointcloud_fusion_camera_ids [default: [0,2,4]]
      • output/tracked_objects [default: /perception/object_recognition/tracking/objects]
      • output/predicted_objects [default: /perception/object_recognition/objects]
      • sync_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/fusion_common.param.yaml]
      • roi_cluster_fusion_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/roi_cluster_fusion.param.yaml]
      • pointpainting_fusion_common_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/pointpainting_common.param.yaml]
      • irregular_object_crop_box_filter_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/crop_box_filter.param.yaml]
      • irregular_object_scan_ground_filter_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/scan_ground_filter.param.yaml]
      • irregular_object_roi_pointcloud_fusion_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/roi_pointcloud_fusion.param.yaml]
      • irregular_object_fusion_camera_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/fusion_cameras.param.yaml]
      • lidar_model_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model]
      • centerpoint_common_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model/centerpoint_common.param.yaml]
      • voxel_grid_based_euclidean_param_path [default: $(var perception_config_dir)/object_recognition/detection/clustering/voxel_grid_based_euclidean_cluster.param.yaml]
      • detection_by_tracker_param_path [default: $(var perception_config_dir)/object_recognition/detection/detection_by_tracker/detection_by_tracker.param.yaml]
      • object_recognition_detection_obstacle_pointcloud_based_validator_param_path [default: $(var perception_config_dir)/object_recognition/detection/detected_object_validation/obstacle_pointcloud_based_validator.param.yaml]
      • object_recognition_detection_object_merger_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/detection/object_merger/data_association_matrix.param.yaml]
      • object_recognition_detection_object_merger_distance_threshold_list_path [default: $(var perception_config_dir)/object_recognition/detection/object_merger/overlapped_judge.param.yaml]
      • ml_camera_lidar_object_association_merger_param_path [default: $(var perception_config_dir)/object_recognition/detection/object_merger/ml_camera_lidar_object_association_merger.param.yaml]
      • roi_detected_object_fusion_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/roi_detected_object_fusion.param.yaml]
      • object_recognition_tracking_multi_object_tracker_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/data_association_matrix.param.yaml]
      • object_recognition_tracking_multi_object_tracker_input_channels_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/input_channels.param.yaml]
      • object_recognition_tracking_multi_object_tracker_node_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/multi_object_tracker_node.param.yaml]
      • object_recognition_tracking_radar_tracked_object_sorter_param_path [default: $(var perception_config_dir)/object_recognition/tracking/radar_object_tracker/tracked_object_sorter.param.yaml]
      • object_recognition_tracking_radar_tracked_object_lanelet_filter_param_path [default: $(var perception_config_dir)/object_recognition/tracking/radar_object_tracker/tracked_object_lanelet_filter.param.yaml]
      • object_recognition_tracking_object_merger_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/tracking/tracking_object_merger/data_association_matrix.param.yaml]
      • object_recognition_tracking_object_merger_node_param_path [default: $(var perception_config_dir)/object_recognition/tracking/tracking_object_merger/decorative_tracker_merger.param.yaml]
      • euclidean_clustering/shape_policy [default: 0]
  • launch/object_recognition/object_recognition_lidar.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • lidar_detection_model [default: centerpoint]
      • use_short_range_detection [default: false]
      • lidar_short_range_detection_model_type [default: centerpoint_short_range]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • bevfusion_model_path [default: $(var data_path)/bevfusion]
      • use_pointcloud_map [default: true]
      • use_validator [default: false]
      • objects_validation_method [default: obstacle_pointcloud]
      • use_object_filter [default: true]
      • objects_filter_method [default: lanelet_filter]
      • use_low_height_cropbox [default: false]
      • use_vector_map [default: true]
      • tracker_publish_merged_objects [default: false]
      • pointcloud_container_name [default: pointcloud_container]
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • input/obstacle_segmentation/pointcloud [default: /perception/obstacle_segmentation/pointcloud]
      • output/tracker_merged_objects [default: /perception/object_recognition/detection/objects]
      • output/tracked_objects [default: /perception/object_recognition/tracking/objects]
      • output/predicted_objects [default: /perception/object_recognition/objects]
      • lidar_model_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model]
      • centerpoint_common_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model/centerpoint_common.param.yaml]
      • voxel_grid_based_euclidean_param_path [default: $(var perception_config_dir)/object_recognition/detection/clustering/voxel_grid_based_euclidean_cluster.param.yaml]
      • object_recognition_detection_obstacle_pointcloud_based_validator_param_path [default: $(var perception_config_dir)/object_recognition/detection/detected_object_validation/obstacle_pointcloud_based_validator.param.yaml]
      • object_recognition_tracking_multi_object_tracker_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/data_association_matrix.param.yaml]
      • object_recognition_tracking_multi_object_tracker_input_channels_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/input_channels.param.yaml]
      • object_recognition_tracking_multi_object_tracker_node_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/multi_object_tracker_node.param.yaml]
  • launch/object_recognition/object_recognition_lidar_ptv3.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • lidar_detection_model [default: centerpoint]
      • use_short_range_detection [default: false]
      • lidar_short_range_detection_model_type [default: centerpoint_short_range]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • bevfusion_model_path [default: $(var data_path)/bevfusion]
      • use_vector_map [default: true]
      • tracker_publish_merged_objects [default: false]
      • pointcloud_container_name [default: pointcloud_container]
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • input/semantic_segmentation/pointcloud [default: /perception/segmented/pointcloud]
      • output/tracker_merged_objects [default: /perception/object_recognition/detection/objects]
      • output/tracked_objects [default: /perception/object_recognition/tracking/objects]
      • output/predicted_objects [default: /perception/object_recognition/objects]
      • lidar_model_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model]
      • centerpoint_common_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model/centerpoint_common.param.yaml]
      • object_recognition_tracking_multi_object_tracker_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/data_association_matrix.param.yaml]
      • object_recognition_tracking_multi_object_tracker_input_channels_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/input_channels.param.yaml]
      • object_recognition_tracking_multi_object_tracker_node_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/multi_object_tracker_node.param.yaml]
      • euclidean_clustering/param_path [default: $(var perception_config_dir)/object_recognition/detection/euclidean_cluster/label_based_euclidean_cluster.param.yaml]
  • launch/object_recognition/object_recognition_lidar_radar.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • lidar_detection_model [default: centerpoint]
      • use_short_range_detection [default: false]
      • lidar_short_range_detection_model_type [default: centerpoint_short_range]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • bevfusion_model_path [default: $(var data_path)/bevfusion]
      • use_low_height_cropbox [default: false]
      • use_vector_map [default: true]
      • tracker_publish_merged_objects [default: false]
      • pointcloud_container_name [default: pointcloud_container]
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • input/semantic_segmentation/pointcloud [default: /perception/segmented/pointcloud]
      • input/radar [default: /sensing/radar/detected_objects]
      • output/tracker_merged_objects [default: /perception/object_recognition/detection/objects]
      • output/tracked_objects [default: /perception/object_recognition/tracking/objects]
      • output/predicted_objects [default: /perception/object_recognition/objects]
      • lidar_model_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model]
      • centerpoint_common_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model/centerpoint_common.param.yaml]
      • voxel_grid_based_euclidean_param_path [default: $(var perception_config_dir)/object_recognition/detection/clustering/voxel_grid_based_euclidean_cluster.param.yaml]
      • radar_lanelet_filtering_range_param_path [default: $(var perception_config_dir)/object_recognition/detection/object_filter/radar_lanelet_filter.param.yaml]
      • object_recognition_detection_object_sorter_radar_param_path [default: $(var perception_config_dir)/object_recognition/detection/radar_filter/detected_object_sorter.param.yaml]
      • object_recognition_tracking_multi_object_tracker_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/data_association_matrix.param.yaml]
      • object_recognition_tracking_multi_object_tracker_input_channels_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/input_channels.param.yaml]
      • object_recognition_tracking_multi_object_tracker_node_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/multi_object_tracker_node.param.yaml]
      • euclidean_clustering/param_path [default: $(var perception_config_dir)/object_recognition/detection/euclidean_cluster/label_based_euclidean_cluster.param.yaml]
  • launch/object_recognition/prediction/prediction.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • use_vector_map [default: false]
      • prediction_model_type [default: map_based]
      • input/objects [default: /perception/object_recognition/tracking/objects]
      • output/objects [default: /perception/object_recognition/objects]
      • object_recognition_prediction_map_based_prediction_param_path [default: $(var perception_config_dir)/object_recognition/prediction/map_based_prediction.param.yaml]
  • launch/obstacle_segmentation/ground_segmentation_additional_lidars.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • target_container
      • lidar_name
      • base_frame [default: base_link]
      • use_intra_process [default: true]
      • input/main/pointcloud
      • input/lidar/pointcloud [default: /sensing/lidar/$(var lidar_name)/pointcloud]
      • output/pointcloud
      • output/pointcloud_info [default: $(var output/pointcloud)_info]
      • crop_box_margin_max_z [default: 2.5]
      • crop_box_margin_min_z [default: -2.5]
      • crop_box_filter_node_param_path [default: $(var perception_config_dir)/obstacle_segmentation/ground_segmentation/crop_box_filter.param.yaml]
      • scan_ground_filter_node_param_path [default: $(var perception_config_dir)/obstacle_segmentation/ground_segmentation/scan_ground_filter.param.yaml]
  • launch/obstacle_segmentation/ground_segmentation_ml.launch.xml
      • pointcloud_container_name [default: pointcloud_container]
      • use_pointcloud_container [default: true]
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • output/pointcloud [default: /perception/obstacle_segmentation/pointcloud]
      • output/pointcloud/segmentation [default: /perception/segmented/pointcloud]
      • output/pointcloud/visualization [default: /perception/segmented/pointcloud/visualization]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • model_variant [default: ptv3]
      • model_path [default: $(var data_path)/ptv3]
  • launch/obstacle_segmentation/ground_segmentation_ransac.launch.xml
      • target_container
      • base_frame [default: base_link]
      • use_intra_process [default: true]
      • input_topics
      • input/main/pointcloud
      • output/pointcloud
      • output/pointcloud_info [default: $(var output/pointcloud)_info]
      • detection_area_crop_box_filter_node_param_path
      • ransac_ground_filter_node_param_path [default: $(find-pkg-share autoware_ground_segmentation)/config/ransac_ground_filter.param.yaml]
  • launch/obstacle_segmentation/ground_segmentation_rule_based.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • target_container
      • base_frame [default: base_link]
      • use_intra_process [default: true]
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • output/pointcloud [default: /perception/obstacle_segmentation/pointcloud]
      • crop_box_margin_max_z [default: 2.5]
      • crop_box_margin_min_z [default: -2.5]
      • crop_box_filter_node_param_path [default: $(var perception_config_dir)/obstacle_segmentation/ground_segmentation/crop_box_filter.param.yaml]
      • scan_ground_filter_node_param_path [default: $(var perception_config_dir)/obstacle_segmentation/ground_segmentation/scan_ground_filter.param.yaml]
  • launch/obstacle_segmentation/ground_segmentation_rule_based_cuda.launch.xml
      • target_container
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • output/pointcloud [default: /perception/obstacle_segmentation/pointcloud]
      • output/ground_pointcloud [default: /perception/obstacle_segmentation/ground_pointcloud]
      • cuda_scan_ground_segmentation_filter_node_param_path [default: $(find-pkg-share autoware_ground_segmentation_cuda)/config/cuda_scan_ground_segmentation_filter.param.yaml]
  • launch/occupancy_grid_map/probabilistic_occupancy_grid_map.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • input/obstacle_pointcloud [default: concatenated/pointcloud]
      • input/raw_pointcloud [default: no_ground/oneshot/pointcloud]
      • output [default: /perception/occupancy_grid_map/map]
      • use_intra_process [default: false]
      • use_multithread [default: false]
      • pointcloud_container_name [default: pointcloud_container]
      • occupancy_grid_map_method
      • occupancy_grid_map_param_path [default: $(var perception_config_dir)/occupancy_grid_map/$(var occupancy_grid_map_method).param.yaml]
      • occupancy_grid_map_updater
      • occupancy_grid_map_updater_param_path [default: $(var perception_config_dir)/occupancy_grid_map/$(var occupancy_grid_map_updater)_updater.param.yaml]
      • input_obstacle_pointcloud [default: false]
      • input_obstacle_and_raw_pointcloud [default: true]
      • use_pointcloud_container [default: true]
  • launch/traffic_light_recognition/traffic_light_detector_fine.launch.xml
    • Fine-detection chain for one traffic light camera: map based detector proposes rough ROIs, the fine detector refines them, classifiers label them.
      • perception_config_pkg [default: autoware_perception_config]
      • camera_namespace
      • enable_image_decompressor [default: true]
      • input/vector_map [default: /map/vector_map]
      • input/route [default: /planning/mission_planning/route]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • map_based_detector_param_path [default: $(find-pkg-share autoware_traffic_light_map_based_detector)/config/traffic_light_map_based_detector.param.yaml]
      • fine_detector_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_fine_detector/traffic_light_fine_detector.param.yaml]
      • car_classifier_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_classifier/car_traffic_light_classifier.param.yaml]
      • pedestrian_classifier_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_classifier/pedestrian_traffic_light_classifier.param.yaml]
      • lamp_recognizer_ml_param_path [default: $(var data_path)/traffic_light_classifier/lamp_recognizer_ml.param.yaml]
      • roi_visualizer_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_visualization/traffic_light_roi_visualizer.param.yaml]
      • fine_detector_model_path [default: $(var data_path)/traffic_light_fine_detector/tlr_car_ped_yolox_s_batch_6.onnx]
      • fine_detector_label_path [default: $(var data_path)/traffic_light_fine_detector/tlr_labels.txt]
      • car_classifier_model_path [default: $(var data_path)/traffic_light_classifier/traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • car_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels.txt]
      • pedestrian_classifier_model_path [default: $(var data_path)/traffic_light_classifier/ped_traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • pedestrian_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels_ped.txt]
      • car_classifier_type [default: 1]
      • pedestrian_classifier_type [default: 1]
  • launch/traffic_light_recognition/traffic_light_detector_rough.launch.xml
    • Rough-detection chain for one traffic light camera: map based detector ROIs feed the classifiers directly, without a high accuracy detector.
      • perception_config_pkg [default: autoware_perception_config]
      • camera_namespace
      • enable_image_decompressor [default: true]
      • input/vector_map [default: /map/vector_map]
      • input/route [default: /planning/mission_planning/route]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • map_based_detector_param_path [default: $(find-pkg-share autoware_traffic_light_map_based_detector)/config/traffic_light_map_based_detector.param.yaml]
      • car_classifier_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_classifier/car_traffic_light_classifier.param.yaml]
      • pedestrian_classifier_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_classifier/pedestrian_traffic_light_classifier.param.yaml]
      • lamp_recognizer_ml_param_path [default: $(var data_path)/traffic_light_classifier/lamp_recognizer_ml.param.yaml]
      • roi_visualizer_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_visualization/traffic_light_roi_visualizer.param.yaml]
      • car_classifier_model_path [default: $(var data_path)/traffic_light_classifier/traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • car_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels.txt]
      • pedestrian_classifier_model_path [default: $(var data_path)/traffic_light_classifier/ped_traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • pedestrian_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels_ped.txt]
      • car_classifier_type [default: 1]
      • pedestrian_classifier_type [default: 1]
  • launch/traffic_light_recognition/traffic_light_detector_whole_image.launch.xml
    • Whole-image detection chain for one traffic light camera: yolox detects over the full image, the selector matches detections to map based ROIs, classifiers label them.
      • perception_config_pkg [default: autoware_perception_config]
      • camera_namespace
      • enable_image_decompressor [default: true]
      • input/vector_map [default: /map/vector_map]
      • input/route [default: /planning/mission_planning/route]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • map_based_detector_param_path [default: $(find-pkg-share autoware_traffic_light_map_based_detector)/config/traffic_light_map_based_detector.param.yaml]
      • whole_image_detector_param_path [default: $(var perception_config_dir)/traffic_light_recognition/tensorrt_yolox/yolox_traffic_light_detector.param.yaml]
      • car_classifier_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_classifier/car_traffic_light_classifier.param.yaml]
      • pedestrian_classifier_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_classifier/pedestrian_traffic_light_classifier.param.yaml]
      • lamp_recognizer_ml_param_path [default: $(var data_path)/traffic_light_classifier/lamp_recognizer_ml.param.yaml]
      • roi_visualizer_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_visualization/traffic_light_roi_visualizer.param.yaml]
      • whole_image_detector_model_path [default: $(var data_path)/tensorrt_yolox/yolox_s_car_ped_tl_detector_960_960_batch_1.onnx]
      • whole_image_detector_label_path [default: $(var data_path)/tensorrt_yolox/car_ped_tl_detector_labels.txt]
      • car_classifier_model_path [default: $(var data_path)/traffic_light_classifier/traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • car_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels.txt]
      • pedestrian_classifier_model_path [default: $(var data_path)/traffic_light_classifier/ped_traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • pedestrian_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels_ped.txt]
      • car_classifier_type [default: 1]
      • pedestrian_classifier_type [default: 1]
  • launch/traffic_light_recognition/traffic_light_fusion.launch.xml
    • Post-process shared by all traffic light cameras: multi camera fusion, V2X arbitration, pedestrian signal estimation and map visualization.
      • perception_config_pkg [default: autoware_perception_config]
      • camera_namespaces [default: [camera6, camera7]]
      • input/vector_map [default: /map/vector_map]
      • external/traffic_signals [default: /perception/traffic_light_recognition/external/traffic_signals]
      • internal/traffic_signals [default: /perception/traffic_light_recognition/internal/traffic_signals]
      • judged/traffic_signals [default: /perception/traffic_light_recognition/judged/traffic_signals]
      • output/traffic_signals [default: /perception/traffic_light_recognition/traffic_signals]
      • multi_camera_fusion_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_multi_camera_fusion/traffic_light_multi_camera_fusion.param.yaml]
      • arbiter_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_arbiter/traffic_light_arbiter.param.yaml]
      • crosswalk_estimator_param_path [default: $(var perception_config_dir)/traffic_light_recognition/crosswalk_traffic_light_estimator/crosswalk_traffic_light_estimator.param.yaml]
  • launch/traffic_light_recognition/traffic_light_occlusion_predictor.launch.xml
    • Occlusion check for one traffic light camera: merges the car and pedestrian signals into the per-camera output, masking signals hidden from the camera.
      • perception_config_pkg [default: autoware_perception_config]
      • camera_namespace
      • input/cloud [default: /sensing/lidar/top/pointcloud_raw_ex]
      • input/vector_map [default: /map/vector_map]
      • occlusion_predictor_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_occlusion_predictor/traffic_light_occlusion_predictor.param.yaml]
  • launch/traffic_light_recognition/traffic_light_recognition_fine.launch.xml
    • Traffic light recognition, fine detection mode: per-camera fine detection with occlusion check, then shared fusion. Two cameras as the representative set; add or remove per-camera blocks to match the vehicle.
      • perception_config_pkg [default: autoware_perception_config]
      • input/vector_map [default: /map/vector_map]
      • input/route [default: /planning/mission_planning/route]
      • input/pointcloud [default: /sensing/lidar/top/pointcloud_raw_ex]
      • external/traffic_signals [default: /perception/traffic_light_recognition/external/traffic_signals]
      • output/traffic_signals [default: /perception/traffic_light_recognition/traffic_signals]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • fine_detector_model_path [default: $(var data_path)/traffic_light_fine_detector/tlr_car_ped_yolox_s_batch_6.onnx]
      • fine_detector_label_path [default: $(var data_path)/traffic_light_fine_detector/tlr_labels.txt]
      • lamp_recognizer_ml_param_path [default: $(var data_path)/traffic_light_classifier/lamp_recognizer_ml.param.yaml]
      • car_classifier_model_path [default: $(var data_path)/traffic_light_classifier/traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • car_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels.txt]
      • pedestrian_classifier_model_path [default: $(var data_path)/traffic_light_classifier/ped_traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • pedestrian_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels_ped.txt]
      • car_classifier_type [default: 1]
      • pedestrian_classifier_type [default: 1]
      • base_map_based_detector_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_map_based_detector/base_traffic_light_map_based_detector.param.yaml]
      • extra_map_based_detector_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_map_based_detector/extra_traffic_light_map_based_detector.param.yaml]
  • launch/traffic_light_recognition/traffic_light_recognition_merge.launch.xml
    • Traffic light recognition, merge mode: detection and classification run on other ECUs; this side runs the per-camera occlusion check and the shared fusion. Two cameras as the representative set; add or remove per-camera blocks to match the vehicle.
      • input/vector_map [default: /map/vector_map]
      • input/pointcloud [default: /sensing/lidar/top/pointcloud_raw_ex]
      • external/traffic_signals [default: /perception/traffic_light_recognition/external/traffic_signals]
      • output/traffic_signals [default: /perception/traffic_light_recognition/traffic_signals]
  • launch/traffic_light_recognition/traffic_light_recognition_rough.launch.xml
    • Traffic light recognition, rough detection mode: per-camera map based detection with occlusion check, then shared fusion. One camera as the representative set; add or remove per-camera blocks to match the vehicle.
      • perception_config_pkg [default: autoware_perception_config]
      • input/vector_map [default: /map/vector_map]
      • input/route [default: /planning/mission_planning/route]
      • input/pointcloud [default: /sensing/lidar/top/pointcloud_raw_ex]
      • external/traffic_signals [default: /perception/traffic_light_recognition/external/traffic_signals]
      • output/traffic_signals [default: /perception/traffic_light_recognition/traffic_signals]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • lamp_recognizer_ml_param_path [default: $(var data_path)/traffic_light_classifier/lamp_recognizer_ml.param.yaml]
      • car_classifier_model_path [default: $(var data_path)/traffic_light_classifier/traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • car_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels.txt]
      • pedestrian_classifier_model_path [default: $(var data_path)/traffic_light_classifier/ped_traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • pedestrian_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels_ped.txt]
      • car_classifier_type [default: 1]
      • pedestrian_classifier_type [default: 1]
      • base_map_based_detector_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_map_based_detector/base_traffic_light_map_based_detector.param.yaml]
  • launch/traffic_light_recognition/traffic_light_recognition_whole_image.launch.xml
    • Traffic light recognition, whole-image detection mode: per-camera yolox detection, then shared fusion. Two cameras as the representative set; add or remove per-camera blocks to match the vehicle.
      • perception_config_pkg [default: autoware_perception_config]
      • input/vector_map [default: /map/vector_map]
      • input/route [default: /planning/mission_planning/route]
      • external/traffic_signals [default: /perception/traffic_light_recognition/external/traffic_signals]
      • output/traffic_signals [default: /perception/traffic_light_recognition/traffic_signals]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • whole_image_detector_model_path [default: $(var data_path)/tensorrt_yolox/yolox_s_car_ped_tl_detector_960_960_batch_1.onnx]
      • whole_image_detector_label_path [default: $(var data_path)/tensorrt_yolox/car_ped_tl_detector_labels.txt]
      • lamp_recognizer_ml_param_path [default: $(var data_path)/traffic_light_classifier/lamp_recognizer_ml.param.yaml]
      • car_classifier_model_path [default: $(var data_path)/traffic_light_classifier/traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • car_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels.txt]
      • pedestrian_classifier_model_path [default: $(var data_path)/traffic_light_classifier/ped_traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • pedestrian_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels_ped.txt]
      • car_classifier_type [default: 1]
      • pedestrian_classifier_type [default: 1]
      • base_map_based_detector_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_map_based_detector/base_traffic_light_map_based_detector.param.yaml]
      • extra_map_based_detector_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_map_based_detector/extra_traffic_light_map_based_detector.param.yaml]

Messages

No message files found.

Services

No service files found

Plugins

No plugins found.

Recent questions tagged autoware_perception_launch at Robotics Stack Exchange

No version for distro galactic showing github. Known supported distros are highlighted in the buttons above.
Package symbol

autoware_perception_launch package from autoware_launch repo

autoware_launch autoware_sample_designs autoware_control_config autoware_control_launch autoware_localization_config autoware_localization_launch autoware_map_config autoware_map_launch autoware_perception_config autoware_perception_launch autoware_planning_config autoware_planning_launch autoware_sensing_launch autoware_sensor_driver_launch awsim_labs_sensor_kit_description awsim_labs_sensor_kit_launch awsim_sensor_kit_description awsim_sensor_kit_launch carla_sensor_kit_description carla_sensor_kit_launch sample_sensor_kit_description sample_sensor_kit_launch single_lidar_sensor_kit_description single_lidar_sensor_kit_launch tier4_autoware_api_launch tier4_simulator_launch tier4_system_launch tier4_vehicle_launch awsim_labs_vehicle_description awsim_labs_vehicle_launch sample_vehicle_description sample_vehicle_launch

ROS Distro
github

API Docs
Browse Code

Package Summary

Version 0.53.0
License Apache License 2.0
Build type AMENT_CMAKE
Use RECOMMENDED

Repository Summary

Description
Checkout URI https://github.com/autowarefoundation/autoware_launch.git
VCS Type git
VCS Version main
Last Updated 2026-10-26
Dev Status UNKNOWN
Released UNRELEASED
Contributing Help Wanted (-)
Good First Issues (-)
Pull Requests to Review (-)

Package Description

The autoware_perception_launch package

Maintainers

  • Yukihiro Saito
  • Dai Nguyen
  • Taekjin Lee
  • Yoshi Ri
  • Masato Saeki

Authors

No additional authors.
autoware_universe_launch/autoware_perception_launch/README.md

autoware_perception_launch

Shared perception launch library. The system component launchers live in the product layer (autoware_launch/launch/components/component_perception.launch.xml for the lidar pipeline and component_traffic_light.launch.xml for traffic light recognition); this package hosts the perception feature launchers they combine:

launch/
├── common/                        # shared preprocess feature launchers (pointcloud downsample)
├── object_recognition/
│   ├── object_recognition_<mode>.launch.xml  # detection + tracking + prediction per sensor mode
│   ├── detection/
│   │   ├── detector/              # detectors
│   │   ├── filter/                # filters
│   │   └── merger/                # detection-side serial merger (used by the camera_lidar_radar_serial launcher)
│   └── prediction/
├── obstacle_segmentation/
├── occupancy_grid_map/
└── traffic_light_recognition/
design/module/                     # Autoware System Designer modules; same subtree as launch/
├── object_recognition/
├── obstacle_segmentation/
├── occupancy_grid_map/
└── traffic_light_recognition/

Launch files

  • object_recognition/object_recognition_<mode>.launch.xml — one feature launcher per sensor mode (camera, lidar, lidar_ptv3, lidar_radar, camera_lidar_radar, camera_lidar_radar_serial); hosts the detection stage inline together with the tracker and prediction, and pushes their module namespaces. Camera and radar input topics default to their absolute sensing paths (/sensing/camera/cameraN/..., /sensing/radar/detected_objects), so callers override only what differs. Most modes wire the multi-channel tracker to the detector outputs; camera_lidar_radar_serial instead merges the detectors in the detection stage (optionally regenerating detections through detection-by-tracker), tracks the single merged detected_objects channel, and joins radar tracked objects at the decorative tracker merger. camera runs camera-only detectors (BEV and camera VRU).
  • The clustering backend is file selection, not a flag: lidar runs rule-based euclidean clustering off the obstacle segmentation cloud, while lidar_ptv3 and lidar_radar cluster the PTv3 semantic-segmentation cloud and drop the pointcloud map filter and the object validator that only the rule-based pipeline uses. Both feed their cluster objects to the tracker on the lidar_clustering channel.
  • object_recognition/detection/{detector,filter}/ — single-purpose launchers wired together by the object-recognition launchers. merger/camera_lidar_merger.launch.xml chains the detection-side object mergers for the camera_lidar_radar_serial launcher.
  • obstacle_segmentation/ground_segmentation_<backend>.launch.xml — ground removal backends (rule_based, rule_based_cuda, ml); backend selection is file selection. ground_segmentation_{additional_lidars,ransac}.launch.xml fold extra lidars into the obstacle stream for kits that carry them.
  • occupancy_grid_map/, common/, traffic_light_recognition/ — feature launchers included directly by the system components.

Namespace and container addressing

Callers own the namespaces: the system component pushes perception and object_recognition; object_recognition_<mode>.launch.xml pushes detection/tracking/prediction. The object-recognition launchers address their internal topics through the fixed /perception/object_recognition namespace. Composable nodes load into the shared pointcloud container, whose name is exposed as pointcloud_container_name and aliased to the node/pointcloud_container argument of the detector launchers.

Config

Parameter files are resolved from a config/ tree addressed through the perception_config_pkg argument, which defaults to autoware_perception_config. A product can relocate the whole tree by setting perception_config_pkg once at its entry point — the launch configuration propagates through every include — but it must then ship the same config/ layout. Products that differ in only a few files instead override the corresponding *_param_path arguments, which every feature launcher declares with a default:

<include file="$(find-pkg-share autoware_perception_launch)/launch/object_recognition/object_recognition_lidar.launch.xml">
  <arg name="perception_config_pkg" value="my_perception_config"/>
  <arg name="FOO_NODE_param_path" value="..."/>
</include>

component_perception.launch.xml, component_traffic_light.launch.xml and planning_simulator/perception_simulation.launch.xml each declare perception_config_pkg, so a product switches the tree at the component it launches.

The launch/ tree is the interface downstream distributions adopt as-is; only package.xml, CMakeLists.txt, README.md, and changelogs stay package-local. Keep changes to this tree folder-merge compatible.

Package Dependencies

Please see <exec_depend> in package.xml.

CHANGELOG

Changelog for package autoware_perception_launch

0.53.0 (2026-10-13)

  • Merge remote-tracking branch 'origin/main' into tmp/bot/bump_version_base

  • feat(autoware_launch): run crop_box_filter standalone when agnocast is enabled (#1990)

    * feat(autoware_launch): run crop_box_filter standalone when agnocast is enabled autoware_universe#13384 moves CropBoxFilterComponent to agnocast_wrapper::Node on an agnocast-only executor, which the component containers do not set up. Start each of the eight crop_box_filter nodes as a standalone process when ENABLE_AGNOCAST=1 and keep loading them into their container otherwise. The node names, namespaces, remaps and parameters are the same in both branches.

    * refactor(autoware_launch): launch crop_box_filter through the autoware_node action ---------

  • feat(autoware_perception_launch): run scan_ground_filter standalone when agnocast is enabled (#1991)

    * feat(autoware_perception_launch): run scan_ground_filter standalone when agnocast is enabled autoware_universe#13385 moves ScanGroundFilterComponent to agnocast_wrapper::Node on an agnocast-only executor, which the component containers do not set up. Start each of the three scan_ground_filter nodes as a standalone process when ENABLE_AGNOCAST=1 and keep loading them into their container otherwise. The node names, namespaces, remaps and parameters are the same in both branches.

    * refactor(autoware_perception_launch): launch scan_ground_filter through the autoware_node action ---------

  • feat(autoware_launch): run voxel_grid_downsample_filter standalone when agnocast is enabled (#1992)

    • feat(autoware_launch): run voxel_grid_downsample_filter standalone when agnocast is enabled
    • style(pre-commit): autofix

    * refactor(autoware_launch): launch voxel_grid_downsample_filter through the autoware_node action ---------Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>

  • feat(autoware_sample_designs): compose the AD API, system and vehicle components from design modules (#1976)

    * fix(control, planning): connect the operation mode state and velocity limit inputs Wires collision_detector's operation_mode_state and remaining_distance_time_calculator's planning_velocity through the module designs; their node designs now declare the fixed topic names as remap targets.

    * fix(planning): connect the internal velocity limit path to the selector Forwards motion_velocity_planner's velocity_limit and clear_velocity_limit up to ScenarioPlanning and feeds them into external_velocity_limit_selector, whose internal inputs were unconnected.

    * fix(control, planning): connect the remaining module inputs to their sources Wires the operation mode transition manager's gate and selector inputs, the stop mode operator's vehicle status, the manual lane change handler's route availability and preferred primitive service, the parking costmap, and the route marker. Exposes the vehicle control mode service, the preferred lane service and the system hazard lights command as component interfaces.

    • fix(planning): follow the hazard lights selector input rename
    • fix(planning): add steering_status subscriber to various planning modules
    • feat(vehicle): add vehicle wrapper and launch file for vehicle interface
    • style(pre-commit): autofix
    • fix(planning): add planning factors publishers and connections for various modules
    • refactor(autoware_sample_designs): compose the AD API from design modules
    • Refactor ADAPI module structure and enhance planning factors
    • Updated AdapiRvizHelpers.module.yaml to include helper prefixes for instances and connections.
    • Modified AutowareSample.system.yaml to reorganize ADAPI components, introducing default_adapi for better structure.
    • Expanded planning factors in Planning.module.yaml, LaneDriving.module.yaml, ScenarioPlanning.module.yaml, and associated modules to include additional factors for improved behavior and motion planning.
    • Enhanced connections in behavior planning and motion planning modules to accommodate new planning factors.
    • Ensured consistency across scenario planning modules by aligning

File truncated at 100 lines see the full file

Package Dependencies

Deps Name
ament_cmake_auto
autoware_cmake
autoware_agnocast_wrapper
autoware_bevfusion
autoware_bytetrack
autoware_compare_map_segmentation
autoware_crosswalk_traffic_light_estimator
autoware_cuda_pointcloud_preprocessor
autoware_detected_object_feature_remover
autoware_detected_object_validation
autoware_detection_by_tracker
autoware_elevation_map_loader
autoware_euclidean_cluster
autoware_ground_segmentation
autoware_ground_segmentation_cuda
autoware_image_object_locator
autoware_image_projection_based_fusion
autoware_image_transport_decompressor
autoware_lidar_apollo_instance_segmentation
autoware_lidar_centerpoint
autoware_lidar_transfusion
autoware_map_based_prediction
autoware_multi_object_tracker
autoware_object_merger
autoware_object_sorter
autoware_occupancy_grid_map_outlier_filter
autoware_perception_config
autoware_pointcloud_preprocessor
autoware_probabilistic_occupancy_grid_map
autoware_ptv3
autoware_raindrop_cluster_filter
autoware_shape_estimation
autoware_simpl_prediction
autoware_simple_object_merger
autoware_tensorrt_bevdet
autoware_tensorrt_yolox
autoware_tracking_object_merger
autoware_traffic_light_arbiter
autoware_traffic_light_category_merger
autoware_traffic_light_classifier
autoware_traffic_light_fine_detector
autoware_traffic_light_map_based_detector
autoware_traffic_light_multi_camera_fusion
autoware_traffic_light_occlusion_predictor
autoware_traffic_light_selector
autoware_traffic_light_visualization
rclcpp_components
topic_tools
ament_lint_auto
autoware_lint_common

System Dependencies

No direct system dependencies.

Dependant Packages

Name Deps
autoware_launch
reaction_analyzer

Launch files

  • launch/common/pointcloud_downsample.launch.xml
      • pointcloud_container_name
      • cuda_pointcloud_preprocessing [default: false]
      • input/pointcloud
      • output/pointcloud
      • common_downsample_voxel_size_x [default: 0.05]
      • common_downsample_voxel_size_y [default: 0.05]
      • common_downsample_voxel_size_z [default: 0.05]
  • launch/object_recognition/detection/detector/camera_2d_detector.launch.xml
      • image_raw0 [default: /sensing/camera/camera0/image_raw]
      • image_raw1 [default: /sensing/camera/camera1/image_raw]
      • image_raw2 [default: /sensing/camera/camera2/image_raw]
      • image_raw3 [default: /sensing/camera/camera3/image_raw]
      • image_raw4 [default: /sensing/camera/camera4/image_raw]
      • image_raw5 [default: /sensing/camera/camera5/image_raw]
      • image_raw6 [default: /sensing/camera/camera6/image_raw]
      • image_raw7 [default: /sensing/camera/camera7/image_raw]
      • image_raw8 [default: /sensing/camera/camera8/image_raw]
      • image_raw9 [default: /sensing/camera/camera9/image_raw]
      • image_number [default: 1]
      • camera_index [default: 0]
      • use_bytetrack [default: true]
      • enable_visualizer [default: false]
      • camera_2d_detector/model_path
      • camera_2d_detector/label_path
      • camera_2d_detector/color_map_path
      • camera_2d_detector/roi_remap_path
      • camera_2d_detector/roi_to_semseg_remap_path
      • tensorrt_yolox_ns [default: ]
  • launch/object_recognition/detection/detector/camera_bev_detector.launch.xml
      • input/camera0/image
      • input/camera0/info
      • input/camera1/image
      • input/camera1/info
      • input/camera2/image
      • input/camera2/info
      • input/camera3/image
      • input/camera3/info
      • input/camera4/image
      • input/camera4/info
      • input/camera5/image
      • input/camera5/info
      • output/objects
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • bevdet_model_name [default: bevdet_one_lt_d]
      • bevdet_model_path [default: $(var data_path)/tensorrt_bevdet]
  • launch/object_recognition/detection/detector/camera_lidar_detector.launch.xml
      • ns
      • lidar_detection_model_type
      • lidar_detection_model_name
      • use_low_intensity_cluster_filter
      • use_image_segmentation_based_filter
      • number_of_cameras
      • input/camera0/image
      • input/camera0/info
      • input/camera0/rois
      • input/camera1/image
      • input/camera1/info
      • input/camera1/rois
      • input/camera2/image
      • input/camera2/info
      • input/camera2/rois
      • input/camera3/image
      • input/camera3/info
      • input/camera3/rois
      • input/camera4/image
      • input/camera4/info
      • input/camera4/rois
      • input/camera5/image
      • input/camera5/info
      • input/camera5/rois
      • input/camera6/image
      • input/camera6/info
      • input/camera6/rois
      • input/camera7/image
      • input/camera7/info
      • input/camera7/rois
      • input/camera8/image
      • input/camera8/info
      • input/camera8/rois
      • input/concatenation_info
      • segmentation_pointcloud_fusion_camera_ids
      • image_topic_name
      • sync_param_path
      • voxel_grid_based_euclidean_param_path
      • lidar_model_param_path
      • pointpainting_model_path [default: $(var data_path)/image_projection_based_fusion]
      • pointpainting_fusion_common_param_path
      • roi_cluster_fusion_param_path
      • node/pointcloud_container
      • input/pointcloud
      • input/pointcloud_map/pointcloud
      • input/obstacle_segmentation/pointcloud
      • output/ml_detector/objects
      • output/rule_detector/objects
      • camera_2d_detector/model_path
      • camera_2d_detector/label_path
      • camera_2d_detector/color_map_path
      • camera_2d_detector/roi_remap_path
      • camera_2d_detector/roi_to_semseg_remap_path
      • enable_2d_detection [default: false]
  • launch/object_recognition/detection/detector/camera_lidar_irregular_object_detector.launch.xml
      • pipeline_ns
      • node/pointcloud_container
      • output/objects [default: objects]
      • input/concatenation_info
      • input/pointcloud
      • irregular_object_crop_box_filter_node_param_path
      • irregular_object_scan_ground_filter_node_param_path
      • irregular_object_roi_pointcloud_fusion_node_param_path
      • irregular_object_fusion_camera_param_path
      • sync_param_path
  • launch/object_recognition/detection/detector/camera_vru_detector.launch.xml
      • input/camera0/info [default: /sensing/camera/camera0/camera_info]
      • input/camera0/rois [default: /perception/object_recognition/detection/rois0]
      • input/camera1/info [default: /sensing/camera/camera1/camera_info]
      • input/camera1/rois [default: /perception/object_recognition/detection/rois1]
      • input/camera2/info [default: /sensing/camera/camera2/camera_info]
      • input/camera2/rois [default: /perception/object_recognition/detection/rois2]
      • input/camera3/info [default: /sensing/camera/camera3/camera_info]
      • input/camera3/rois [default: /perception/object_recognition/detection/rois3]
      • input/camera4/info [default: /sensing/camera/camera4/camera_info]
      • input/camera4/rois [default: /perception/object_recognition/detection/rois4]
      • input/camera5/info [default: /sensing/camera/camera5/camera_info]
      • input/camera5/rois [default: /perception/object_recognition/detection/rois5]
      • input/camera6/info [default: /sensing/camera/camera6/camera_info]
      • input/camera6/rois [default: /perception/object_recognition/detection/rois6]
      • input/camera7/info [default: /sensing/camera/camera7/camera_info]
      • input/camera7/rois [default: /perception/object_recognition/detection/rois7]
      • output/objects [default: /perception/object_recognition/detection/camera_vru/objects]
      • bbox_object_locator_param_path
      • rois_ids [default: [0, 1]]
  • launch/object_recognition/detection/detector/irregular_object_detector.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • target_container
      • base_frame [default: base_link]
      • use_intra_process [default: true]
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • input/concatenation_info [default: /sensing/lidar/concatenated/pointcloud_info]
      • output/clusters
      • crop_box_filter_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/crop_box_filter.param.yaml]
      • scan_ground_filter_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/scan_ground_filter.param.yaml]
      • roi_pointcloud_fusion_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/roi_pointcloud_fusion.param.yaml]
      • fusion_camera_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/fusion_cameras.param.yaml]
      • sync_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/fusion_common.param.yaml]
  • launch/object_recognition/detection/detector/lidar_dnn_detector.launch.xml
      • lidar_detection_model_type
      • lidar_detection_model_name
      • lidar_model_param_path
      • bevfusion_model_path [default: $(var data_path)/bevfusion]
      • centerpoint_model_path [default: $(var data_path)/lidar_centerpoint]
      • centerpoint_common_param_path [default: $(find-pkg-share autoware_lidar_centerpoint)/config/centerpoint_common.param.yaml]
      • transfusion_model_path [default: $(var data_path)/lidar_transfusion]
      • use_short_range_detection [default: false]
      • lidar_short_range_detection_model_type
      • node/pointcloud_container
      • input/pointcloud
      • output/objects
      • output/short_range_objects
  • launch/object_recognition/detection/detector/lidar_rule_detector.launch.xml
      • ns
      • node/pointcloud_container
      • input/pointcloud_map/pointcloud
      • input/obstacle_segmentation/pointcloud
      • output/cluster_objects
      • output/objects
      • voxel_grid_based_euclidean_param_path
      • use_low_height_cropbox [default: false]
  • launch/object_recognition/detection/detector/lidar_semantic_segmentation_detector.launch.xml
      • euclidean_clustering/param_path
      • input/pointcloud
      • output/objects
  • launch/object_recognition/detection/detector/tracker_based_detector.launch.xml
      • detection_by_tracker_param_path
      • input/clusters
      • input/tracked_objects
      • output/objects
  • launch/object_recognition/detection/filter/object_filter.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • objects_filter_method [default: lanelet_filter]
      • object_recognition_detection_object_lanelet_filter_param_path [default: $(var perception_config_dir)/object_recognition/detection/object_filter/object_lanelet_filter.param.yaml]
      • object_recognition_detection_object_position_filter_param_path [default: $(var perception_config_dir)/object_recognition/detection/object_filter/object_position_filter.param.yaml]
      • input/objects
      • output/objects
  • launch/object_recognition/detection/filter/object_validator.launch.xml
      • objects_validation_method
      • object_recognition_detection_obstacle_pointcloud_based_validator_param_path
      • input/obstacle_pointcloud
      • input/objects
      • output/objects
  • launch/object_recognition/detection/filter/pointcloud_downsample_filter.launch.xml
      • target_container
      • use_intra_process [default: true]
      • input/pointcloud
      • output/pointcloud
      • downsample_voxel_size [default: 0.1]
  • launch/object_recognition/detection/filter/pointcloud_map_filter.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • target_container
      • use_intra_process [default: true]
      • input/pointcloud
      • output/pointcloud
      • output/downsampled/pointcloud [default: /perception/obstacle_segmentation/pointcloud_map_filtered/downsampled/pointcloud]
      • downsample_voxel_size [default: 0.1]
      • voxel_based_compare_map_filter_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/pointcloud_filter/voxel_based_compare_map_filter.param.yaml]
  • launch/object_recognition/detection/filter/radar_filter.launch.xml
      • object_sorter_param_path
      • radar_lanelet_filtering_range_param_path
      • input/radar
      • output/objects
  • launch/object_recognition/detection/merger/camera_lidar_merger.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • object_recognition_detection_object_merger_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/detection/object_merger/data_association_matrix.param.yaml]
      • object_recognition_detection_object_merger_distance_threshold_list_path [default: $(var perception_config_dir)/object_recognition/detection/object_merger/overlapped_judge.param.yaml]
      • ml_camera_lidar_object_association_merger_param_path [default: $(var perception_config_dir)/object_recognition/detection/object_merger/ml_camera_lidar_object_association_merger.param.yaml]
      • roi_detected_object_fusion_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/roi_detected_object_fusion.param.yaml]
      • sync_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/fusion_common.param.yaml]
      • lidar_detection_model_type
      • use_detection_by_tracker
      • use_irregular_object_detector
      • use_roi_detected_object_validator
      • euclidean_clustering/shape_policy
      • use_object_filter
      • objects_filter_method
      • number_of_cameras
      • input/camera0/image
      • input/camera0/info
      • input/camera0/rois
      • input/camera1/image
      • input/camera1/info
      • input/camera1/rois
      • input/camera2/image
      • input/camera2/info
      • input/camera2/rois
      • input/camera3/image
      • input/camera3/info
      • input/camera3/rois
      • input/camera4/image
      • input/camera4/info
      • input/camera4/rois
      • input/camera5/image
      • input/camera5/info
      • input/camera5/rois
      • input/camera6/image
      • input/camera6/info
      • input/camera6/rois
      • input/camera7/image
      • input/camera7/info
      • input/camera7/rois
      • input/camera8/image
      • input/camera8/info
      • input/camera8/rois
      • input/concatenation_info [default: /sensing/lidar/concatenated/pointcloud_info]
      • input/lidar_ml/objects
      • input/lidar_cluster/objects
      • input/irregular_object/objects [default: ]
      • input/detection_by_tracker/objects
      • output/objects [default: objects]
      • alpha_merger_priority_mode [default: 0]
  • launch/object_recognition/object_recognition_camera.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • use_camera_vru_detector [default: false]
      • camera_vru_detector_rois_ids [default: [0]]
      • use_camera_bev_detector [default: true]
      • use_vector_map [default: true]
      • tracker_publish_merged_objects [default: false]
      • input/camera0/image [default: /sensing/camera/camera0/image_rect_color]
      • input/camera0/info [default: /sensing/camera/camera0/camera_info]
      • input/camera0/rois [default: /perception/object_recognition/detection/rois0]
      • input/camera1/image [default: /sensing/camera/camera1/image_rect_color]
      • input/camera1/info [default: /sensing/camera/camera1/camera_info]
      • input/camera1/rois [default: /perception/object_recognition/detection/rois1]
      • input/camera2/image [default: /sensing/camera/camera2/image_rect_color]
      • input/camera2/info [default: /sensing/camera/camera2/camera_info]
      • input/camera2/rois [default: /perception/object_recognition/detection/rois2]
      • input/camera3/image [default: /sensing/camera/camera3/image_rect_color]
      • input/camera3/info [default: /sensing/camera/camera3/camera_info]
      • input/camera3/rois [default: /perception/object_recognition/detection/rois3]
      • input/camera4/image [default: /sensing/camera/camera4/image_rect_color]
      • input/camera4/info [default: /sensing/camera/camera4/camera_info]
      • input/camera4/rois [default: /perception/object_recognition/detection/rois4]
      • input/camera5/image [default: /sensing/camera/camera5/image_rect_color]
      • input/camera5/info [default: /sensing/camera/camera5/camera_info]
      • input/camera5/rois [default: /perception/object_recognition/detection/rois5]
      • input/camera6/image [default: /sensing/camera/camera6/image_rect_color]
      • input/camera6/info [default: /sensing/camera/camera6/camera_info]
      • input/camera6/rois [default: /perception/object_recognition/detection/rois6]
      • input/camera7/image [default: /sensing/camera/camera7/image_rect_color]
      • input/camera7/info [default: /sensing/camera/camera7/camera_info]
      • input/camera7/rois [default: /perception/object_recognition/detection/rois7]
      • output/tracker_merged_objects [default: /perception/object_recognition/detection/objects]
      • output/tracked_objects [default: /perception/object_recognition/tracking/objects]
      • output/predicted_objects [default: /perception/object_recognition/objects]
      • near_range_camera_vru_param_path [default: $(var perception_config_dir)/object_recognition/detection/camera_vru_detection/near_range_camera_vru_detector.param.yaml]
      • object_recognition_tracking_multi_object_tracker_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/data_association_matrix.param.yaml]
      • object_recognition_tracking_multi_object_tracker_input_channels_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/input_channels.param.yaml]
      • object_recognition_tracking_multi_object_tracker_node_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/multi_object_tracker_node.param.yaml]
  • launch/object_recognition/object_recognition_camera_lidar_radar.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • lidar_detection_model [default: centerpoint]
      • use_short_range_detection [default: false]
      • lidar_short_range_detection_model_type [default: centerpoint_short_range]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • bevfusion_model_path [default: $(var data_path)/bevfusion]
      • pointpainting_model_path [default: $(var data_path)/image_projection_based_fusion]
      • camera_2d_detector/model_path [default: $(var data_path)/tensorrt_yolox/yolox-sPlus-opt-pseudoV2-T4-960x960-T4-seg16cls.onnx]
      • camera_2d_detector/label_path [default: $(var data_path)/tensorrt_yolox/label.txt]
      • camera_2d_detector/color_map_path [default: $(var data_path)/tensorrt_yolox/semseg_color_map.csv]
      • camera_2d_detector/roi_remap_path [default: $(var perception_config_dir)/traffic_light_recognition/tensorrt_yolox/roi_label_remap.csv]
      • camera_2d_detector/roi_to_semseg_remap_path [default: $(var perception_config_dir)/traffic_light_recognition/tensorrt_yolox/roi_to_semseg_label_remap.csv]
      • use_semantic_segmentation_ptv3 [default: false]
      • use_object_filter [default: true]
      • use_pointcloud_map [default: true]
      • use_validator [default: false]
      • objects_validation_method [default: obstacle_pointcloud]
      • use_low_height_cropbox [default: false]
      • use_low_intensity_cluster_filter [default: true]
      • use_image_segmentation_based_filter [default: false]
      • use_irregular_object_detector [default: true]
      • use_camera_vru_detector [default: false]
      • camera_vru_detector_rois_ids [default: [0]]
      • use_camera_only_detection [default: false]
      • use_vector_map [default: true]
      • tracker_publish_merged_objects [default: false]
      • number_of_cameras [default: 6]
      • pointcloud_container_name [default: pointcloud_container]
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • input/semantic_segmentation/pointcloud [default: /perception/segmented/pointcloud]
      • input/obstacle_segmentation/pointcloud [default: /perception/obstacle_segmentation/pointcloud]
      • input/camera0/image [default: /sensing/camera/camera0/image_rect_color]
      • input/camera0/info [default: /sensing/camera/camera0/camera_info]
      • input/camera0/rois [default: /perception/object_recognition/detection/rois0]
      • input/camera1/image [default: /sensing/camera/camera1/image_rect_color]
      • input/camera1/info [default: /sensing/camera/camera1/camera_info]
      • input/camera1/rois [default: /perception/object_recognition/detection/rois1]
      • input/camera2/image [default: /sensing/camera/camera2/image_rect_color]
      • input/camera2/info [default: /sensing/camera/camera2/camera_info]
      • input/camera2/rois [default: /perception/object_recognition/detection/rois2]
      • input/camera3/image [default: /sensing/camera/camera3/image_rect_color]
      • input/camera3/info [default: /sensing/camera/camera3/camera_info]
      • input/camera3/rois [default: /perception/object_recognition/detection/rois3]
      • input/camera4/image [default: /sensing/camera/camera4/image_rect_color]
      • input/camera4/info [default: /sensing/camera/camera4/camera_info]
      • input/camera4/rois [default: /perception/object_recognition/detection/rois4]
      • input/camera5/image [default: /sensing/camera/camera5/image_rect_color]
      • input/camera5/info [default: /sensing/camera/camera5/camera_info]
      • input/camera5/rois [default: /perception/object_recognition/detection/rois5]
      • input/camera6/image [default: /sensing/camera/camera6/image_rect_color]
      • input/camera6/info [default: /sensing/camera/camera6/camera_info]
      • input/camera6/rois [default: /perception/object_recognition/detection/rois6]
      • input/camera7/image [default: /sensing/camera/camera7/image_rect_color]
      • input/camera7/info [default: /sensing/camera/camera7/camera_info]
      • input/camera7/rois [default: /perception/object_recognition/detection/rois7]
      • input/camera8/image [default: /sensing/camera/camera8/image_rect_color]
      • input/camera8/info [default: /sensing/camera/camera8/camera_info]
      • input/camera8/rois [default: /perception/object_recognition/detection/rois8]
      • input/concatenation_info [default: /sensing/lidar/concatenated/pointcloud_info]
      • input/radar [default: /sensing/radar/detected_objects]
      • input/camera_only/objects [default: /perception/object_recognition/detection/camera_only/objects]
      • input/camera_only/channel [default: camera_streampetr]
      • image_topic_name [default: image_rect_color]
      • segmentation_pointcloud_fusion_camera_ids [default: [0,2,4]]
      • output/tracker_merged_objects [default: /perception/object_recognition/detection/objects]
      • output/tracked_objects [default: /perception/object_recognition/tracking/objects]
      • output/predicted_objects [default: /perception/object_recognition/objects]
      • sync_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/fusion_common.param.yaml]
      • roi_cluster_fusion_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/roi_cluster_fusion.param.yaml]
      • pointpainting_fusion_common_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/pointpainting_common.param.yaml]
      • irregular_object_crop_box_filter_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/crop_box_filter.param.yaml]
      • irregular_object_scan_ground_filter_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/scan_ground_filter.param.yaml]
      • irregular_object_roi_pointcloud_fusion_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/roi_pointcloud_fusion.param.yaml]
      • irregular_object_fusion_camera_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/fusion_cameras.param.yaml]
      • lidar_model_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model]
      • centerpoint_common_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model/centerpoint_common.param.yaml]
      • voxel_grid_based_euclidean_param_path [default: $(var perception_config_dir)/object_recognition/detection/clustering/voxel_grid_based_euclidean_cluster.param.yaml]
      • radar_lanelet_filtering_range_param_path [default: $(var perception_config_dir)/object_recognition/detection/object_filter/radar_lanelet_filter.param.yaml]
      • object_recognition_detection_object_sorter_radar_param_path [default: $(var perception_config_dir)/object_recognition/detection/radar_filter/detected_object_sorter.param.yaml]
      • near_range_camera_vru_param_path [default: $(var perception_config_dir)/object_recognition/detection/camera_vru_detection/near_range_camera_vru_detector.param.yaml]
      • object_recognition_detection_obstacle_pointcloud_based_validator_param_path [default: $(var perception_config_dir)/object_recognition/detection/detected_object_validation/obstacle_pointcloud_based_validator.param.yaml]
      • object_recognition_tracking_multi_object_tracker_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/data_association_matrix.param.yaml]
      • object_recognition_tracking_multi_object_tracker_input_channels_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/input_channels.param.yaml]
      • object_recognition_tracking_multi_object_tracker_node_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/multi_object_tracker_node.param.yaml]
      • euclidean_clustering/param_path [default: $(var perception_config_dir)/object_recognition/detection/euclidean_cluster/label_based_euclidean_cluster.param.yaml]
  • launch/object_recognition/object_recognition_camera_lidar_radar_serial.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • lidar_detection_model [default: centerpoint]
      • use_short_range_detection [default: false]
      • lidar_short_range_detection_model_type [default: centerpoint_short_range]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • bevfusion_model_path [default: $(var data_path)/bevfusion]
      • pointpainting_model_path [default: $(var data_path)/image_projection_based_fusion]
      • camera_2d_detector/model_path [default: $(var data_path)/tensorrt_yolox/yolox-sPlus-opt-pseudoV2-T4-960x960-T4-seg16cls.onnx]
      • camera_2d_detector/label_path [default: $(var data_path)/tensorrt_yolox/label.txt]
      • camera_2d_detector/color_map_path [default: $(var data_path)/tensorrt_yolox/semseg_color_map.csv]
      • camera_2d_detector/roi_remap_path [default: $(var perception_config_dir)/traffic_light_recognition/tensorrt_yolox/roi_label_remap.csv]
      • camera_2d_detector/roi_to_semseg_remap_path [default: $(var perception_config_dir)/traffic_light_recognition/tensorrt_yolox/roi_to_semseg_label_remap.csv]
      • use_object_filter [default: true]
      • objects_filter_method [default: lanelet_filter]
      • use_pointcloud_map [default: true]
      • use_validator [default: false]
      • objects_validation_method [default: obstacle_pointcloud]
      • use_roi_detected_object_validator [default: true]
      • use_detection_by_tracker [default: true]
      • use_low_height_cropbox [default: false]
      • use_low_intensity_cluster_filter [default: true]
      • use_image_segmentation_based_filter [default: false]
      • use_irregular_object_detector [default: true]
      • ml_camera_lidar_merger_priority_mode [default: 0]
      • use_vector_map [default: true]
      • number_of_cameras [default: 6]
      • pointcloud_container_name [default: pointcloud_container]
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • input/obstacle_segmentation/pointcloud [default: /perception/obstacle_segmentation/pointcloud]
      • input/camera0/image [default: /sensing/camera/camera0/image_rect_color]
      • input/camera0/info [default: /sensing/camera/camera0/camera_info]
      • input/camera0/rois [default: /perception/object_recognition/detection/rois0]
      • input/camera1/image [default: /sensing/camera/camera1/image_rect_color]
      • input/camera1/info [default: /sensing/camera/camera1/camera_info]
      • input/camera1/rois [default: /perception/object_recognition/detection/rois1]
      • input/camera2/image [default: /sensing/camera/camera2/image_rect_color]
      • input/camera2/info [default: /sensing/camera/camera2/camera_info]
      • input/camera2/rois [default: /perception/object_recognition/detection/rois2]
      • input/camera3/image [default: /sensing/camera/camera3/image_rect_color]
      • input/camera3/info [default: /sensing/camera/camera3/camera_info]
      • input/camera3/rois [default: /perception/object_recognition/detection/rois3]
      • input/camera4/image [default: /sensing/camera/camera4/image_rect_color]
      • input/camera4/info [default: /sensing/camera/camera4/camera_info]
      • input/camera4/rois [default: /perception/object_recognition/detection/rois4]
      • input/camera5/image [default: /sensing/camera/camera5/image_rect_color]
      • input/camera5/info [default: /sensing/camera/camera5/camera_info]
      • input/camera5/rois [default: /perception/object_recognition/detection/rois5]
      • input/camera6/image [default: /sensing/camera/camera6/image_rect_color]
      • input/camera6/info [default: /sensing/camera/camera6/camera_info]
      • input/camera6/rois [default: /perception/object_recognition/detection/rois6]
      • input/camera7/image [default: /sensing/camera/camera7/image_rect_color]
      • input/camera7/info [default: /sensing/camera/camera7/camera_info]
      • input/camera7/rois [default: /perception/object_recognition/detection/rois7]
      • input/camera8/image [default: /sensing/camera/camera8/image_rect_color]
      • input/camera8/info [default: /sensing/camera/camera8/camera_info]
      • input/camera8/rois [default: /perception/object_recognition/detection/rois8]
      • input/concatenation_info [default: /sensing/lidar/concatenated/pointcloud_info]
      • input/radar/tracked_objects [default: /sensing/radar/tracked_objects]
      • image_topic_name [default: image_rect_color]
      • segmentation_pointcloud_fusion_camera_ids [default: [0,2,4]]
      • output/tracked_objects [default: /perception/object_recognition/tracking/objects]
      • output/predicted_objects [default: /perception/object_recognition/objects]
      • sync_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/fusion_common.param.yaml]
      • roi_cluster_fusion_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/roi_cluster_fusion.param.yaml]
      • pointpainting_fusion_common_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/pointpainting_common.param.yaml]
      • irregular_object_crop_box_filter_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/crop_box_filter.param.yaml]
      • irregular_object_scan_ground_filter_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/scan_ground_filter.param.yaml]
      • irregular_object_roi_pointcloud_fusion_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/roi_pointcloud_fusion.param.yaml]
      • irregular_object_fusion_camera_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/fusion_cameras.param.yaml]
      • lidar_model_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model]
      • centerpoint_common_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model/centerpoint_common.param.yaml]
      • voxel_grid_based_euclidean_param_path [default: $(var perception_config_dir)/object_recognition/detection/clustering/voxel_grid_based_euclidean_cluster.param.yaml]
      • detection_by_tracker_param_path [default: $(var perception_config_dir)/object_recognition/detection/detection_by_tracker/detection_by_tracker.param.yaml]
      • object_recognition_detection_obstacle_pointcloud_based_validator_param_path [default: $(var perception_config_dir)/object_recognition/detection/detected_object_validation/obstacle_pointcloud_based_validator.param.yaml]
      • object_recognition_detection_object_merger_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/detection/object_merger/data_association_matrix.param.yaml]
      • object_recognition_detection_object_merger_distance_threshold_list_path [default: $(var perception_config_dir)/object_recognition/detection/object_merger/overlapped_judge.param.yaml]
      • ml_camera_lidar_object_association_merger_param_path [default: $(var perception_config_dir)/object_recognition/detection/object_merger/ml_camera_lidar_object_association_merger.param.yaml]
      • roi_detected_object_fusion_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/roi_detected_object_fusion.param.yaml]
      • object_recognition_tracking_multi_object_tracker_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/data_association_matrix.param.yaml]
      • object_recognition_tracking_multi_object_tracker_input_channels_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/input_channels.param.yaml]
      • object_recognition_tracking_multi_object_tracker_node_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/multi_object_tracker_node.param.yaml]
      • object_recognition_tracking_radar_tracked_object_sorter_param_path [default: $(var perception_config_dir)/object_recognition/tracking/radar_object_tracker/tracked_object_sorter.param.yaml]
      • object_recognition_tracking_radar_tracked_object_lanelet_filter_param_path [default: $(var perception_config_dir)/object_recognition/tracking/radar_object_tracker/tracked_object_lanelet_filter.param.yaml]
      • object_recognition_tracking_object_merger_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/tracking/tracking_object_merger/data_association_matrix.param.yaml]
      • object_recognition_tracking_object_merger_node_param_path [default: $(var perception_config_dir)/object_recognition/tracking/tracking_object_merger/decorative_tracker_merger.param.yaml]
      • euclidean_clustering/shape_policy [default: 0]
  • launch/object_recognition/object_recognition_lidar.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • lidar_detection_model [default: centerpoint]
      • use_short_range_detection [default: false]
      • lidar_short_range_detection_model_type [default: centerpoint_short_range]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • bevfusion_model_path [default: $(var data_path)/bevfusion]
      • use_pointcloud_map [default: true]
      • use_validator [default: false]
      • objects_validation_method [default: obstacle_pointcloud]
      • use_object_filter [default: true]
      • objects_filter_method [default: lanelet_filter]
      • use_low_height_cropbox [default: false]
      • use_vector_map [default: true]
      • tracker_publish_merged_objects [default: false]
      • pointcloud_container_name [default: pointcloud_container]
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • input/obstacle_segmentation/pointcloud [default: /perception/obstacle_segmentation/pointcloud]
      • output/tracker_merged_objects [default: /perception/object_recognition/detection/objects]
      • output/tracked_objects [default: /perception/object_recognition/tracking/objects]
      • output/predicted_objects [default: /perception/object_recognition/objects]
      • lidar_model_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model]
      • centerpoint_common_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model/centerpoint_common.param.yaml]
      • voxel_grid_based_euclidean_param_path [default: $(var perception_config_dir)/object_recognition/detection/clustering/voxel_grid_based_euclidean_cluster.param.yaml]
      • object_recognition_detection_obstacle_pointcloud_based_validator_param_path [default: $(var perception_config_dir)/object_recognition/detection/detected_object_validation/obstacle_pointcloud_based_validator.param.yaml]
      • object_recognition_tracking_multi_object_tracker_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/data_association_matrix.param.yaml]
      • object_recognition_tracking_multi_object_tracker_input_channels_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/input_channels.param.yaml]
      • object_recognition_tracking_multi_object_tracker_node_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/multi_object_tracker_node.param.yaml]
  • launch/object_recognition/object_recognition_lidar_ptv3.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • lidar_detection_model [default: centerpoint]
      • use_short_range_detection [default: false]
      • lidar_short_range_detection_model_type [default: centerpoint_short_range]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • bevfusion_model_path [default: $(var data_path)/bevfusion]
      • use_vector_map [default: true]
      • tracker_publish_merged_objects [default: false]
      • pointcloud_container_name [default: pointcloud_container]
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • input/semantic_segmentation/pointcloud [default: /perception/segmented/pointcloud]
      • output/tracker_merged_objects [default: /perception/object_recognition/detection/objects]
      • output/tracked_objects [default: /perception/object_recognition/tracking/objects]
      • output/predicted_objects [default: /perception/object_recognition/objects]
      • lidar_model_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model]
      • centerpoint_common_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model/centerpoint_common.param.yaml]
      • object_recognition_tracking_multi_object_tracker_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/data_association_matrix.param.yaml]
      • object_recognition_tracking_multi_object_tracker_input_channels_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/input_channels.param.yaml]
      • object_recognition_tracking_multi_object_tracker_node_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/multi_object_tracker_node.param.yaml]
      • euclidean_clustering/param_path [default: $(var perception_config_dir)/object_recognition/detection/euclidean_cluster/label_based_euclidean_cluster.param.yaml]
  • launch/object_recognition/object_recognition_lidar_radar.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • lidar_detection_model [default: centerpoint]
      • use_short_range_detection [default: false]
      • lidar_short_range_detection_model_type [default: centerpoint_short_range]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • bevfusion_model_path [default: $(var data_path)/bevfusion]
      • use_low_height_cropbox [default: false]
      • use_vector_map [default: true]
      • tracker_publish_merged_objects [default: false]
      • pointcloud_container_name [default: pointcloud_container]
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • input/semantic_segmentation/pointcloud [default: /perception/segmented/pointcloud]
      • input/radar [default: /sensing/radar/detected_objects]
      • output/tracker_merged_objects [default: /perception/object_recognition/detection/objects]
      • output/tracked_objects [default: /perception/object_recognition/tracking/objects]
      • output/predicted_objects [default: /perception/object_recognition/objects]
      • lidar_model_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model]
      • centerpoint_common_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model/centerpoint_common.param.yaml]
      • voxel_grid_based_euclidean_param_path [default: $(var perception_config_dir)/object_recognition/detection/clustering/voxel_grid_based_euclidean_cluster.param.yaml]
      • radar_lanelet_filtering_range_param_path [default: $(var perception_config_dir)/object_recognition/detection/object_filter/radar_lanelet_filter.param.yaml]
      • object_recognition_detection_object_sorter_radar_param_path [default: $(var perception_config_dir)/object_recognition/detection/radar_filter/detected_object_sorter.param.yaml]
      • object_recognition_tracking_multi_object_tracker_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/data_association_matrix.param.yaml]
      • object_recognition_tracking_multi_object_tracker_input_channels_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/input_channels.param.yaml]
      • object_recognition_tracking_multi_object_tracker_node_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/multi_object_tracker_node.param.yaml]
      • euclidean_clustering/param_path [default: $(var perception_config_dir)/object_recognition/detection/euclidean_cluster/label_based_euclidean_cluster.param.yaml]
  • launch/object_recognition/prediction/prediction.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • use_vector_map [default: false]
      • prediction_model_type [default: map_based]
      • input/objects [default: /perception/object_recognition/tracking/objects]
      • output/objects [default: /perception/object_recognition/objects]
      • object_recognition_prediction_map_based_prediction_param_path [default: $(var perception_config_dir)/object_recognition/prediction/map_based_prediction.param.yaml]
  • launch/obstacle_segmentation/ground_segmentation_additional_lidars.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • target_container
      • lidar_name
      • base_frame [default: base_link]
      • use_intra_process [default: true]
      • input/main/pointcloud
      • input/lidar/pointcloud [default: /sensing/lidar/$(var lidar_name)/pointcloud]
      • output/pointcloud
      • output/pointcloud_info [default: $(var output/pointcloud)_info]
      • crop_box_margin_max_z [default: 2.5]
      • crop_box_margin_min_z [default: -2.5]
      • crop_box_filter_node_param_path [default: $(var perception_config_dir)/obstacle_segmentation/ground_segmentation/crop_box_filter.param.yaml]
      • scan_ground_filter_node_param_path [default: $(var perception_config_dir)/obstacle_segmentation/ground_segmentation/scan_ground_filter.param.yaml]
  • launch/obstacle_segmentation/ground_segmentation_ml.launch.xml
      • pointcloud_container_name [default: pointcloud_container]
      • use_pointcloud_container [default: true]
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • output/pointcloud [default: /perception/obstacle_segmentation/pointcloud]
      • output/pointcloud/segmentation [default: /perception/segmented/pointcloud]
      • output/pointcloud/visualization [default: /perception/segmented/pointcloud/visualization]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • model_variant [default: ptv3]
      • model_path [default: $(var data_path)/ptv3]
  • launch/obstacle_segmentation/ground_segmentation_ransac.launch.xml
      • target_container
      • base_frame [default: base_link]
      • use_intra_process [default: true]
      • input_topics
      • input/main/pointcloud
      • output/pointcloud
      • output/pointcloud_info [default: $(var output/pointcloud)_info]
      • detection_area_crop_box_filter_node_param_path
      • ransac_ground_filter_node_param_path [default: $(find-pkg-share autoware_ground_segmentation)/config/ransac_ground_filter.param.yaml]
  • launch/obstacle_segmentation/ground_segmentation_rule_based.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • target_container
      • base_frame [default: base_link]
      • use_intra_process [default: true]
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • output/pointcloud [default: /perception/obstacle_segmentation/pointcloud]
      • crop_box_margin_max_z [default: 2.5]
      • crop_box_margin_min_z [default: -2.5]
      • crop_box_filter_node_param_path [default: $(var perception_config_dir)/obstacle_segmentation/ground_segmentation/crop_box_filter.param.yaml]
      • scan_ground_filter_node_param_path [default: $(var perception_config_dir)/obstacle_segmentation/ground_segmentation/scan_ground_filter.param.yaml]
  • launch/obstacle_segmentation/ground_segmentation_rule_based_cuda.launch.xml
      • target_container
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • output/pointcloud [default: /perception/obstacle_segmentation/pointcloud]
      • output/ground_pointcloud [default: /perception/obstacle_segmentation/ground_pointcloud]
      • cuda_scan_ground_segmentation_filter_node_param_path [default: $(find-pkg-share autoware_ground_segmentation_cuda)/config/cuda_scan_ground_segmentation_filter.param.yaml]
  • launch/occupancy_grid_map/probabilistic_occupancy_grid_map.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • input/obstacle_pointcloud [default: concatenated/pointcloud]
      • input/raw_pointcloud [default: no_ground/oneshot/pointcloud]
      • output [default: /perception/occupancy_grid_map/map]
      • use_intra_process [default: false]
      • use_multithread [default: false]
      • pointcloud_container_name [default: pointcloud_container]
      • occupancy_grid_map_method
      • occupancy_grid_map_param_path [default: $(var perception_config_dir)/occupancy_grid_map/$(var occupancy_grid_map_method).param.yaml]
      • occupancy_grid_map_updater
      • occupancy_grid_map_updater_param_path [default: $(var perception_config_dir)/occupancy_grid_map/$(var occupancy_grid_map_updater)_updater.param.yaml]
      • input_obstacle_pointcloud [default: false]
      • input_obstacle_and_raw_pointcloud [default: true]
      • use_pointcloud_container [default: true]
  • launch/traffic_light_recognition/traffic_light_detector_fine.launch.xml
    • Fine-detection chain for one traffic light camera: map based detector proposes rough ROIs, the fine detector refines them, classifiers label them.
      • perception_config_pkg [default: autoware_perception_config]
      • camera_namespace
      • enable_image_decompressor [default: true]
      • input/vector_map [default: /map/vector_map]
      • input/route [default: /planning/mission_planning/route]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • map_based_detector_param_path [default: $(find-pkg-share autoware_traffic_light_map_based_detector)/config/traffic_light_map_based_detector.param.yaml]
      • fine_detector_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_fine_detector/traffic_light_fine_detector.param.yaml]
      • car_classifier_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_classifier/car_traffic_light_classifier.param.yaml]
      • pedestrian_classifier_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_classifier/pedestrian_traffic_light_classifier.param.yaml]
      • lamp_recognizer_ml_param_path [default: $(var data_path)/traffic_light_classifier/lamp_recognizer_ml.param.yaml]
      • roi_visualizer_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_visualization/traffic_light_roi_visualizer.param.yaml]
      • fine_detector_model_path [default: $(var data_path)/traffic_light_fine_detector/tlr_car_ped_yolox_s_batch_6.onnx]
      • fine_detector_label_path [default: $(var data_path)/traffic_light_fine_detector/tlr_labels.txt]
      • car_classifier_model_path [default: $(var data_path)/traffic_light_classifier/traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • car_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels.txt]
      • pedestrian_classifier_model_path [default: $(var data_path)/traffic_light_classifier/ped_traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • pedestrian_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels_ped.txt]
      • car_classifier_type [default: 1]
      • pedestrian_classifier_type [default: 1]
  • launch/traffic_light_recognition/traffic_light_detector_rough.launch.xml
    • Rough-detection chain for one traffic light camera: map based detector ROIs feed the classifiers directly, without a high accuracy detector.
      • perception_config_pkg [default: autoware_perception_config]
      • camera_namespace
      • enable_image_decompressor [default: true]
      • input/vector_map [default: /map/vector_map]
      • input/route [default: /planning/mission_planning/route]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • map_based_detector_param_path [default: $(find-pkg-share autoware_traffic_light_map_based_detector)/config/traffic_light_map_based_detector.param.yaml]
      • car_classifier_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_classifier/car_traffic_light_classifier.param.yaml]
      • pedestrian_classifier_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_classifier/pedestrian_traffic_light_classifier.param.yaml]
      • lamp_recognizer_ml_param_path [default: $(var data_path)/traffic_light_classifier/lamp_recognizer_ml.param.yaml]
      • roi_visualizer_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_visualization/traffic_light_roi_visualizer.param.yaml]
      • car_classifier_model_path [default: $(var data_path)/traffic_light_classifier/traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • car_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels.txt]
      • pedestrian_classifier_model_path [default: $(var data_path)/traffic_light_classifier/ped_traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • pedestrian_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels_ped.txt]
      • car_classifier_type [default: 1]
      • pedestrian_classifier_type [default: 1]
  • launch/traffic_light_recognition/traffic_light_detector_whole_image.launch.xml
    • Whole-image detection chain for one traffic light camera: yolox detects over the full image, the selector matches detections to map based ROIs, classifiers label them.
      • perception_config_pkg [default: autoware_perception_config]
      • camera_namespace
      • enable_image_decompressor [default: true]
      • input/vector_map [default: /map/vector_map]
      • input/route [default: /planning/mission_planning/route]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • map_based_detector_param_path [default: $(find-pkg-share autoware_traffic_light_map_based_detector)/config/traffic_light_map_based_detector.param.yaml]
      • whole_image_detector_param_path [default: $(var perception_config_dir)/traffic_light_recognition/tensorrt_yolox/yolox_traffic_light_detector.param.yaml]
      • car_classifier_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_classifier/car_traffic_light_classifier.param.yaml]
      • pedestrian_classifier_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_classifier/pedestrian_traffic_light_classifier.param.yaml]
      • lamp_recognizer_ml_param_path [default: $(var data_path)/traffic_light_classifier/lamp_recognizer_ml.param.yaml]
      • roi_visualizer_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_visualization/traffic_light_roi_visualizer.param.yaml]
      • whole_image_detector_model_path [default: $(var data_path)/tensorrt_yolox/yolox_s_car_ped_tl_detector_960_960_batch_1.onnx]
      • whole_image_detector_label_path [default: $(var data_path)/tensorrt_yolox/car_ped_tl_detector_labels.txt]
      • car_classifier_model_path [default: $(var data_path)/traffic_light_classifier/traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • car_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels.txt]
      • pedestrian_classifier_model_path [default: $(var data_path)/traffic_light_classifier/ped_traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • pedestrian_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels_ped.txt]
      • car_classifier_type [default: 1]
      • pedestrian_classifier_type [default: 1]
  • launch/traffic_light_recognition/traffic_light_fusion.launch.xml
    • Post-process shared by all traffic light cameras: multi camera fusion, V2X arbitration, pedestrian signal estimation and map visualization.
      • perception_config_pkg [default: autoware_perception_config]
      • camera_namespaces [default: [camera6, camera7]]
      • input/vector_map [default: /map/vector_map]
      • external/traffic_signals [default: /perception/traffic_light_recognition/external/traffic_signals]
      • internal/traffic_signals [default: /perception/traffic_light_recognition/internal/traffic_signals]
      • judged/traffic_signals [default: /perception/traffic_light_recognition/judged/traffic_signals]
      • output/traffic_signals [default: /perception/traffic_light_recognition/traffic_signals]
      • multi_camera_fusion_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_multi_camera_fusion/traffic_light_multi_camera_fusion.param.yaml]
      • arbiter_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_arbiter/traffic_light_arbiter.param.yaml]
      • crosswalk_estimator_param_path [default: $(var perception_config_dir)/traffic_light_recognition/crosswalk_traffic_light_estimator/crosswalk_traffic_light_estimator.param.yaml]
  • launch/traffic_light_recognition/traffic_light_occlusion_predictor.launch.xml
    • Occlusion check for one traffic light camera: merges the car and pedestrian signals into the per-camera output, masking signals hidden from the camera.
      • perception_config_pkg [default: autoware_perception_config]
      • camera_namespace
      • input/cloud [default: /sensing/lidar/top/pointcloud_raw_ex]
      • input/vector_map [default: /map/vector_map]
      • occlusion_predictor_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_occlusion_predictor/traffic_light_occlusion_predictor.param.yaml]
  • launch/traffic_light_recognition/traffic_light_recognition_fine.launch.xml
    • Traffic light recognition, fine detection mode: per-camera fine detection with occlusion check, then shared fusion. Two cameras as the representative set; add or remove per-camera blocks to match the vehicle.
      • perception_config_pkg [default: autoware_perception_config]
      • input/vector_map [default: /map/vector_map]
      • input/route [default: /planning/mission_planning/route]
      • input/pointcloud [default: /sensing/lidar/top/pointcloud_raw_ex]
      • external/traffic_signals [default: /perception/traffic_light_recognition/external/traffic_signals]
      • output/traffic_signals [default: /perception/traffic_light_recognition/traffic_signals]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • fine_detector_model_path [default: $(var data_path)/traffic_light_fine_detector/tlr_car_ped_yolox_s_batch_6.onnx]
      • fine_detector_label_path [default: $(var data_path)/traffic_light_fine_detector/tlr_labels.txt]
      • lamp_recognizer_ml_param_path [default: $(var data_path)/traffic_light_classifier/lamp_recognizer_ml.param.yaml]
      • car_classifier_model_path [default: $(var data_path)/traffic_light_classifier/traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • car_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels.txt]
      • pedestrian_classifier_model_path [default: $(var data_path)/traffic_light_classifier/ped_traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • pedestrian_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels_ped.txt]
      • car_classifier_type [default: 1]
      • pedestrian_classifier_type [default: 1]
      • base_map_based_detector_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_map_based_detector/base_traffic_light_map_based_detector.param.yaml]
      • extra_map_based_detector_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_map_based_detector/extra_traffic_light_map_based_detector.param.yaml]
  • launch/traffic_light_recognition/traffic_light_recognition_merge.launch.xml
    • Traffic light recognition, merge mode: detection and classification run on other ECUs; this side runs the per-camera occlusion check and the shared fusion. Two cameras as the representative set; add or remove per-camera blocks to match the vehicle.
      • input/vector_map [default: /map/vector_map]
      • input/pointcloud [default: /sensing/lidar/top/pointcloud_raw_ex]
      • external/traffic_signals [default: /perception/traffic_light_recognition/external/traffic_signals]
      • output/traffic_signals [default: /perception/traffic_light_recognition/traffic_signals]
  • launch/traffic_light_recognition/traffic_light_recognition_rough.launch.xml
    • Traffic light recognition, rough detection mode: per-camera map based detection with occlusion check, then shared fusion. One camera as the representative set; add or remove per-camera blocks to match the vehicle.
      • perception_config_pkg [default: autoware_perception_config]
      • input/vector_map [default: /map/vector_map]
      • input/route [default: /planning/mission_planning/route]
      • input/pointcloud [default: /sensing/lidar/top/pointcloud_raw_ex]
      • external/traffic_signals [default: /perception/traffic_light_recognition/external/traffic_signals]
      • output/traffic_signals [default: /perception/traffic_light_recognition/traffic_signals]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • lamp_recognizer_ml_param_path [default: $(var data_path)/traffic_light_classifier/lamp_recognizer_ml.param.yaml]
      • car_classifier_model_path [default: $(var data_path)/traffic_light_classifier/traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • car_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels.txt]
      • pedestrian_classifier_model_path [default: $(var data_path)/traffic_light_classifier/ped_traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • pedestrian_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels_ped.txt]
      • car_classifier_type [default: 1]
      • pedestrian_classifier_type [default: 1]
      • base_map_based_detector_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_map_based_detector/base_traffic_light_map_based_detector.param.yaml]
  • launch/traffic_light_recognition/traffic_light_recognition_whole_image.launch.xml
    • Traffic light recognition, whole-image detection mode: per-camera yolox detection, then shared fusion. Two cameras as the representative set; add or remove per-camera blocks to match the vehicle.
      • perception_config_pkg [default: autoware_perception_config]
      • input/vector_map [default: /map/vector_map]
      • input/route [default: /planning/mission_planning/route]
      • external/traffic_signals [default: /perception/traffic_light_recognition/external/traffic_signals]
      • output/traffic_signals [default: /perception/traffic_light_recognition/traffic_signals]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • whole_image_detector_model_path [default: $(var data_path)/tensorrt_yolox/yolox_s_car_ped_tl_detector_960_960_batch_1.onnx]
      • whole_image_detector_label_path [default: $(var data_path)/tensorrt_yolox/car_ped_tl_detector_labels.txt]
      • lamp_recognizer_ml_param_path [default: $(var data_path)/traffic_light_classifier/lamp_recognizer_ml.param.yaml]
      • car_classifier_model_path [default: $(var data_path)/traffic_light_classifier/traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • car_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels.txt]
      • pedestrian_classifier_model_path [default: $(var data_path)/traffic_light_classifier/ped_traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • pedestrian_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels_ped.txt]
      • car_classifier_type [default: 1]
      • pedestrian_classifier_type [default: 1]
      • base_map_based_detector_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_map_based_detector/base_traffic_light_map_based_detector.param.yaml]
      • extra_map_based_detector_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_map_based_detector/extra_traffic_light_map_based_detector.param.yaml]

Messages

No message files found.

Services

No service files found

Plugins

No plugins found.

Recent questions tagged autoware_perception_launch at Robotics Stack Exchange

No version for distro iron showing github. Known supported distros are highlighted in the buttons above.
Package symbol

autoware_perception_launch package from autoware_launch repo

autoware_launch autoware_sample_designs autoware_control_config autoware_control_launch autoware_localization_config autoware_localization_launch autoware_map_config autoware_map_launch autoware_perception_config autoware_perception_launch autoware_planning_config autoware_planning_launch autoware_sensing_launch autoware_sensor_driver_launch awsim_labs_sensor_kit_description awsim_labs_sensor_kit_launch awsim_sensor_kit_description awsim_sensor_kit_launch carla_sensor_kit_description carla_sensor_kit_launch sample_sensor_kit_description sample_sensor_kit_launch single_lidar_sensor_kit_description single_lidar_sensor_kit_launch tier4_autoware_api_launch tier4_simulator_launch tier4_system_launch tier4_vehicle_launch awsim_labs_vehicle_description awsim_labs_vehicle_launch sample_vehicle_description sample_vehicle_launch

ROS Distro
github

API Docs
Browse Code

Package Summary

Version 0.53.0
License Apache License 2.0
Build type AMENT_CMAKE
Use RECOMMENDED

Repository Summary

Description
Checkout URI https://github.com/autowarefoundation/autoware_launch.git
VCS Type git
VCS Version main
Last Updated 2026-10-26
Dev Status UNKNOWN
Released UNRELEASED
Contributing Help Wanted (-)
Good First Issues (-)
Pull Requests to Review (-)

Package Description

The autoware_perception_launch package

Maintainers

  • Yukihiro Saito
  • Dai Nguyen
  • Taekjin Lee
  • Yoshi Ri
  • Masato Saeki

Authors

No additional authors.
autoware_universe_launch/autoware_perception_launch/README.md

autoware_perception_launch

Shared perception launch library. The system component launchers live in the product layer (autoware_launch/launch/components/component_perception.launch.xml for the lidar pipeline and component_traffic_light.launch.xml for traffic light recognition); this package hosts the perception feature launchers they combine:

launch/
├── common/                        # shared preprocess feature launchers (pointcloud downsample)
├── object_recognition/
│   ├── object_recognition_<mode>.launch.xml  # detection + tracking + prediction per sensor mode
│   ├── detection/
│   │   ├── detector/              # detectors
│   │   ├── filter/                # filters
│   │   └── merger/                # detection-side serial merger (used by the camera_lidar_radar_serial launcher)
│   └── prediction/
├── obstacle_segmentation/
├── occupancy_grid_map/
└── traffic_light_recognition/
design/module/                     # Autoware System Designer modules; same subtree as launch/
├── object_recognition/
├── obstacle_segmentation/
├── occupancy_grid_map/
└── traffic_light_recognition/

Launch files

  • object_recognition/object_recognition_<mode>.launch.xml — one feature launcher per sensor mode (camera, lidar, lidar_ptv3, lidar_radar, camera_lidar_radar, camera_lidar_radar_serial); hosts the detection stage inline together with the tracker and prediction, and pushes their module namespaces. Camera and radar input topics default to their absolute sensing paths (/sensing/camera/cameraN/..., /sensing/radar/detected_objects), so callers override only what differs. Most modes wire the multi-channel tracker to the detector outputs; camera_lidar_radar_serial instead merges the detectors in the detection stage (optionally regenerating detections through detection-by-tracker), tracks the single merged detected_objects channel, and joins radar tracked objects at the decorative tracker merger. camera runs camera-only detectors (BEV and camera VRU).
  • The clustering backend is file selection, not a flag: lidar runs rule-based euclidean clustering off the obstacle segmentation cloud, while lidar_ptv3 and lidar_radar cluster the PTv3 semantic-segmentation cloud and drop the pointcloud map filter and the object validator that only the rule-based pipeline uses. Both feed their cluster objects to the tracker on the lidar_clustering channel.
  • object_recognition/detection/{detector,filter}/ — single-purpose launchers wired together by the object-recognition launchers. merger/camera_lidar_merger.launch.xml chains the detection-side object mergers for the camera_lidar_radar_serial launcher.
  • obstacle_segmentation/ground_segmentation_<backend>.launch.xml — ground removal backends (rule_based, rule_based_cuda, ml); backend selection is file selection. ground_segmentation_{additional_lidars,ransac}.launch.xml fold extra lidars into the obstacle stream for kits that carry them.
  • occupancy_grid_map/, common/, traffic_light_recognition/ — feature launchers included directly by the system components.

Namespace and container addressing

Callers own the namespaces: the system component pushes perception and object_recognition; object_recognition_<mode>.launch.xml pushes detection/tracking/prediction. The object-recognition launchers address their internal topics through the fixed /perception/object_recognition namespace. Composable nodes load into the shared pointcloud container, whose name is exposed as pointcloud_container_name and aliased to the node/pointcloud_container argument of the detector launchers.

Config

Parameter files are resolved from a config/ tree addressed through the perception_config_pkg argument, which defaults to autoware_perception_config. A product can relocate the whole tree by setting perception_config_pkg once at its entry point — the launch configuration propagates through every include — but it must then ship the same config/ layout. Products that differ in only a few files instead override the corresponding *_param_path arguments, which every feature launcher declares with a default:

<include file="$(find-pkg-share autoware_perception_launch)/launch/object_recognition/object_recognition_lidar.launch.xml">
  <arg name="perception_config_pkg" value="my_perception_config"/>
  <arg name="FOO_NODE_param_path" value="..."/>
</include>

component_perception.launch.xml, component_traffic_light.launch.xml and planning_simulator/perception_simulation.launch.xml each declare perception_config_pkg, so a product switches the tree at the component it launches.

The launch/ tree is the interface downstream distributions adopt as-is; only package.xml, CMakeLists.txt, README.md, and changelogs stay package-local. Keep changes to this tree folder-merge compatible.

Package Dependencies

Please see <exec_depend> in package.xml.

CHANGELOG

Changelog for package autoware_perception_launch

0.53.0 (2026-10-13)

  • Merge remote-tracking branch 'origin/main' into tmp/bot/bump_version_base

  • feat(autoware_launch): run crop_box_filter standalone when agnocast is enabled (#1990)

    * feat(autoware_launch): run crop_box_filter standalone when agnocast is enabled autoware_universe#13384 moves CropBoxFilterComponent to agnocast_wrapper::Node on an agnocast-only executor, which the component containers do not set up. Start each of the eight crop_box_filter nodes as a standalone process when ENABLE_AGNOCAST=1 and keep loading them into their container otherwise. The node names, namespaces, remaps and parameters are the same in both branches.

    * refactor(autoware_launch): launch crop_box_filter through the autoware_node action ---------

  • feat(autoware_perception_launch): run scan_ground_filter standalone when agnocast is enabled (#1991)

    * feat(autoware_perception_launch): run scan_ground_filter standalone when agnocast is enabled autoware_universe#13385 moves ScanGroundFilterComponent to agnocast_wrapper::Node on an agnocast-only executor, which the component containers do not set up. Start each of the three scan_ground_filter nodes as a standalone process when ENABLE_AGNOCAST=1 and keep loading them into their container otherwise. The node names, namespaces, remaps and parameters are the same in both branches.

    * refactor(autoware_perception_launch): launch scan_ground_filter through the autoware_node action ---------

  • feat(autoware_launch): run voxel_grid_downsample_filter standalone when agnocast is enabled (#1992)

    • feat(autoware_launch): run voxel_grid_downsample_filter standalone when agnocast is enabled
    • style(pre-commit): autofix

    * refactor(autoware_launch): launch voxel_grid_downsample_filter through the autoware_node action ---------Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>

  • feat(autoware_sample_designs): compose the AD API, system and vehicle components from design modules (#1976)

    * fix(control, planning): connect the operation mode state and velocity limit inputs Wires collision_detector's operation_mode_state and remaining_distance_time_calculator's planning_velocity through the module designs; their node designs now declare the fixed topic names as remap targets.

    * fix(planning): connect the internal velocity limit path to the selector Forwards motion_velocity_planner's velocity_limit and clear_velocity_limit up to ScenarioPlanning and feeds them into external_velocity_limit_selector, whose internal inputs were unconnected.

    * fix(control, planning): connect the remaining module inputs to their sources Wires the operation mode transition manager's gate and selector inputs, the stop mode operator's vehicle status, the manual lane change handler's route availability and preferred primitive service, the parking costmap, and the route marker. Exposes the vehicle control mode service, the preferred lane service and the system hazard lights command as component interfaces.

    • fix(planning): follow the hazard lights selector input rename
    • fix(planning): add steering_status subscriber to various planning modules
    • feat(vehicle): add vehicle wrapper and launch file for vehicle interface
    • style(pre-commit): autofix
    • fix(planning): add planning factors publishers and connections for various modules
    • refactor(autoware_sample_designs): compose the AD API from design modules
    • Refactor ADAPI module structure and enhance planning factors
    • Updated AdapiRvizHelpers.module.yaml to include helper prefixes for instances and connections.
    • Modified AutowareSample.system.yaml to reorganize ADAPI components, introducing default_adapi for better structure.
    • Expanded planning factors in Planning.module.yaml, LaneDriving.module.yaml, ScenarioPlanning.module.yaml, and associated modules to include additional factors for improved behavior and motion planning.
    • Enhanced connections in behavior planning and motion planning modules to accommodate new planning factors.
    • Ensured consistency across scenario planning modules by aligning

File truncated at 100 lines see the full file

Package Dependencies

Deps Name
ament_cmake_auto
autoware_cmake
autoware_agnocast_wrapper
autoware_bevfusion
autoware_bytetrack
autoware_compare_map_segmentation
autoware_crosswalk_traffic_light_estimator
autoware_cuda_pointcloud_preprocessor
autoware_detected_object_feature_remover
autoware_detected_object_validation
autoware_detection_by_tracker
autoware_elevation_map_loader
autoware_euclidean_cluster
autoware_ground_segmentation
autoware_ground_segmentation_cuda
autoware_image_object_locator
autoware_image_projection_based_fusion
autoware_image_transport_decompressor
autoware_lidar_apollo_instance_segmentation
autoware_lidar_centerpoint
autoware_lidar_transfusion
autoware_map_based_prediction
autoware_multi_object_tracker
autoware_object_merger
autoware_object_sorter
autoware_occupancy_grid_map_outlier_filter
autoware_perception_config
autoware_pointcloud_preprocessor
autoware_probabilistic_occupancy_grid_map
autoware_ptv3
autoware_raindrop_cluster_filter
autoware_shape_estimation
autoware_simpl_prediction
autoware_simple_object_merger
autoware_tensorrt_bevdet
autoware_tensorrt_yolox
autoware_tracking_object_merger
autoware_traffic_light_arbiter
autoware_traffic_light_category_merger
autoware_traffic_light_classifier
autoware_traffic_light_fine_detector
autoware_traffic_light_map_based_detector
autoware_traffic_light_multi_camera_fusion
autoware_traffic_light_occlusion_predictor
autoware_traffic_light_selector
autoware_traffic_light_visualization
rclcpp_components
topic_tools
ament_lint_auto
autoware_lint_common

System Dependencies

No direct system dependencies.

Dependant Packages

Name Deps
autoware_launch
reaction_analyzer

Launch files

  • launch/common/pointcloud_downsample.launch.xml
      • pointcloud_container_name
      • cuda_pointcloud_preprocessing [default: false]
      • input/pointcloud
      • output/pointcloud
      • common_downsample_voxel_size_x [default: 0.05]
      • common_downsample_voxel_size_y [default: 0.05]
      • common_downsample_voxel_size_z [default: 0.05]
  • launch/object_recognition/detection/detector/camera_2d_detector.launch.xml
      • image_raw0 [default: /sensing/camera/camera0/image_raw]
      • image_raw1 [default: /sensing/camera/camera1/image_raw]
      • image_raw2 [default: /sensing/camera/camera2/image_raw]
      • image_raw3 [default: /sensing/camera/camera3/image_raw]
      • image_raw4 [default: /sensing/camera/camera4/image_raw]
      • image_raw5 [default: /sensing/camera/camera5/image_raw]
      • image_raw6 [default: /sensing/camera/camera6/image_raw]
      • image_raw7 [default: /sensing/camera/camera7/image_raw]
      • image_raw8 [default: /sensing/camera/camera8/image_raw]
      • image_raw9 [default: /sensing/camera/camera9/image_raw]
      • image_number [default: 1]
      • camera_index [default: 0]
      • use_bytetrack [default: true]
      • enable_visualizer [default: false]
      • camera_2d_detector/model_path
      • camera_2d_detector/label_path
      • camera_2d_detector/color_map_path
      • camera_2d_detector/roi_remap_path
      • camera_2d_detector/roi_to_semseg_remap_path
      • tensorrt_yolox_ns [default: ]
  • launch/object_recognition/detection/detector/camera_bev_detector.launch.xml
      • input/camera0/image
      • input/camera0/info
      • input/camera1/image
      • input/camera1/info
      • input/camera2/image
      • input/camera2/info
      • input/camera3/image
      • input/camera3/info
      • input/camera4/image
      • input/camera4/info
      • input/camera5/image
      • input/camera5/info
      • output/objects
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • bevdet_model_name [default: bevdet_one_lt_d]
      • bevdet_model_path [default: $(var data_path)/tensorrt_bevdet]
  • launch/object_recognition/detection/detector/camera_lidar_detector.launch.xml
      • ns
      • lidar_detection_model_type
      • lidar_detection_model_name
      • use_low_intensity_cluster_filter
      • use_image_segmentation_based_filter
      • number_of_cameras
      • input/camera0/image
      • input/camera0/info
      • input/camera0/rois
      • input/camera1/image
      • input/camera1/info
      • input/camera1/rois
      • input/camera2/image
      • input/camera2/info
      • input/camera2/rois
      • input/camera3/image
      • input/camera3/info
      • input/camera3/rois
      • input/camera4/image
      • input/camera4/info
      • input/camera4/rois
      • input/camera5/image
      • input/camera5/info
      • input/camera5/rois
      • input/camera6/image
      • input/camera6/info
      • input/camera6/rois
      • input/camera7/image
      • input/camera7/info
      • input/camera7/rois
      • input/camera8/image
      • input/camera8/info
      • input/camera8/rois
      • input/concatenation_info
      • segmentation_pointcloud_fusion_camera_ids
      • image_topic_name
      • sync_param_path
      • voxel_grid_based_euclidean_param_path
      • lidar_model_param_path
      • pointpainting_model_path [default: $(var data_path)/image_projection_based_fusion]
      • pointpainting_fusion_common_param_path
      • roi_cluster_fusion_param_path
      • node/pointcloud_container
      • input/pointcloud
      • input/pointcloud_map/pointcloud
      • input/obstacle_segmentation/pointcloud
      • output/ml_detector/objects
      • output/rule_detector/objects
      • camera_2d_detector/model_path
      • camera_2d_detector/label_path
      • camera_2d_detector/color_map_path
      • camera_2d_detector/roi_remap_path
      • camera_2d_detector/roi_to_semseg_remap_path
      • enable_2d_detection [default: false]
  • launch/object_recognition/detection/detector/camera_lidar_irregular_object_detector.launch.xml
      • pipeline_ns
      • node/pointcloud_container
      • output/objects [default: objects]
      • input/concatenation_info
      • input/pointcloud
      • irregular_object_crop_box_filter_node_param_path
      • irregular_object_scan_ground_filter_node_param_path
      • irregular_object_roi_pointcloud_fusion_node_param_path
      • irregular_object_fusion_camera_param_path
      • sync_param_path
  • launch/object_recognition/detection/detector/camera_vru_detector.launch.xml
      • input/camera0/info [default: /sensing/camera/camera0/camera_info]
      • input/camera0/rois [default: /perception/object_recognition/detection/rois0]
      • input/camera1/info [default: /sensing/camera/camera1/camera_info]
      • input/camera1/rois [default: /perception/object_recognition/detection/rois1]
      • input/camera2/info [default: /sensing/camera/camera2/camera_info]
      • input/camera2/rois [default: /perception/object_recognition/detection/rois2]
      • input/camera3/info [default: /sensing/camera/camera3/camera_info]
      • input/camera3/rois [default: /perception/object_recognition/detection/rois3]
      • input/camera4/info [default: /sensing/camera/camera4/camera_info]
      • input/camera4/rois [default: /perception/object_recognition/detection/rois4]
      • input/camera5/info [default: /sensing/camera/camera5/camera_info]
      • input/camera5/rois [default: /perception/object_recognition/detection/rois5]
      • input/camera6/info [default: /sensing/camera/camera6/camera_info]
      • input/camera6/rois [default: /perception/object_recognition/detection/rois6]
      • input/camera7/info [default: /sensing/camera/camera7/camera_info]
      • input/camera7/rois [default: /perception/object_recognition/detection/rois7]
      • output/objects [default: /perception/object_recognition/detection/camera_vru/objects]
      • bbox_object_locator_param_path
      • rois_ids [default: [0, 1]]
  • launch/object_recognition/detection/detector/irregular_object_detector.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • target_container
      • base_frame [default: base_link]
      • use_intra_process [default: true]
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • input/concatenation_info [default: /sensing/lidar/concatenated/pointcloud_info]
      • output/clusters
      • crop_box_filter_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/crop_box_filter.param.yaml]
      • scan_ground_filter_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/scan_ground_filter.param.yaml]
      • roi_pointcloud_fusion_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/roi_pointcloud_fusion.param.yaml]
      • fusion_camera_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/fusion_cameras.param.yaml]
      • sync_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/fusion_common.param.yaml]
  • launch/object_recognition/detection/detector/lidar_dnn_detector.launch.xml
      • lidar_detection_model_type
      • lidar_detection_model_name
      • lidar_model_param_path
      • bevfusion_model_path [default: $(var data_path)/bevfusion]
      • centerpoint_model_path [default: $(var data_path)/lidar_centerpoint]
      • centerpoint_common_param_path [default: $(find-pkg-share autoware_lidar_centerpoint)/config/centerpoint_common.param.yaml]
      • transfusion_model_path [default: $(var data_path)/lidar_transfusion]
      • use_short_range_detection [default: false]
      • lidar_short_range_detection_model_type
      • node/pointcloud_container
      • input/pointcloud
      • output/objects
      • output/short_range_objects
  • launch/object_recognition/detection/detector/lidar_rule_detector.launch.xml
      • ns
      • node/pointcloud_container
      • input/pointcloud_map/pointcloud
      • input/obstacle_segmentation/pointcloud
      • output/cluster_objects
      • output/objects
      • voxel_grid_based_euclidean_param_path
      • use_low_height_cropbox [default: false]
  • launch/object_recognition/detection/detector/lidar_semantic_segmentation_detector.launch.xml
      • euclidean_clustering/param_path
      • input/pointcloud
      • output/objects
  • launch/object_recognition/detection/detector/tracker_based_detector.launch.xml
      • detection_by_tracker_param_path
      • input/clusters
      • input/tracked_objects
      • output/objects
  • launch/object_recognition/detection/filter/object_filter.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • objects_filter_method [default: lanelet_filter]
      • object_recognition_detection_object_lanelet_filter_param_path [default: $(var perception_config_dir)/object_recognition/detection/object_filter/object_lanelet_filter.param.yaml]
      • object_recognition_detection_object_position_filter_param_path [default: $(var perception_config_dir)/object_recognition/detection/object_filter/object_position_filter.param.yaml]
      • input/objects
      • output/objects
  • launch/object_recognition/detection/filter/object_validator.launch.xml
      • objects_validation_method
      • object_recognition_detection_obstacle_pointcloud_based_validator_param_path
      • input/obstacle_pointcloud
      • input/objects
      • output/objects
  • launch/object_recognition/detection/filter/pointcloud_downsample_filter.launch.xml
      • target_container
      • use_intra_process [default: true]
      • input/pointcloud
      • output/pointcloud
      • downsample_voxel_size [default: 0.1]
  • launch/object_recognition/detection/filter/pointcloud_map_filter.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • target_container
      • use_intra_process [default: true]
      • input/pointcloud
      • output/pointcloud
      • output/downsampled/pointcloud [default: /perception/obstacle_segmentation/pointcloud_map_filtered/downsampled/pointcloud]
      • downsample_voxel_size [default: 0.1]
      • voxel_based_compare_map_filter_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/pointcloud_filter/voxel_based_compare_map_filter.param.yaml]
  • launch/object_recognition/detection/filter/radar_filter.launch.xml
      • object_sorter_param_path
      • radar_lanelet_filtering_range_param_path
      • input/radar
      • output/objects
  • launch/object_recognition/detection/merger/camera_lidar_merger.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • object_recognition_detection_object_merger_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/detection/object_merger/data_association_matrix.param.yaml]
      • object_recognition_detection_object_merger_distance_threshold_list_path [default: $(var perception_config_dir)/object_recognition/detection/object_merger/overlapped_judge.param.yaml]
      • ml_camera_lidar_object_association_merger_param_path [default: $(var perception_config_dir)/object_recognition/detection/object_merger/ml_camera_lidar_object_association_merger.param.yaml]
      • roi_detected_object_fusion_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/roi_detected_object_fusion.param.yaml]
      • sync_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/fusion_common.param.yaml]
      • lidar_detection_model_type
      • use_detection_by_tracker
      • use_irregular_object_detector
      • use_roi_detected_object_validator
      • euclidean_clustering/shape_policy
      • use_object_filter
      • objects_filter_method
      • number_of_cameras
      • input/camera0/image
      • input/camera0/info
      • input/camera0/rois
      • input/camera1/image
      • input/camera1/info
      • input/camera1/rois
      • input/camera2/image
      • input/camera2/info
      • input/camera2/rois
      • input/camera3/image
      • input/camera3/info
      • input/camera3/rois
      • input/camera4/image
      • input/camera4/info
      • input/camera4/rois
      • input/camera5/image
      • input/camera5/info
      • input/camera5/rois
      • input/camera6/image
      • input/camera6/info
      • input/camera6/rois
      • input/camera7/image
      • input/camera7/info
      • input/camera7/rois
      • input/camera8/image
      • input/camera8/info
      • input/camera8/rois
      • input/concatenation_info [default: /sensing/lidar/concatenated/pointcloud_info]
      • input/lidar_ml/objects
      • input/lidar_cluster/objects
      • input/irregular_object/objects [default: ]
      • input/detection_by_tracker/objects
      • output/objects [default: objects]
      • alpha_merger_priority_mode [default: 0]
  • launch/object_recognition/object_recognition_camera.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • use_camera_vru_detector [default: false]
      • camera_vru_detector_rois_ids [default: [0]]
      • use_camera_bev_detector [default: true]
      • use_vector_map [default: true]
      • tracker_publish_merged_objects [default: false]
      • input/camera0/image [default: /sensing/camera/camera0/image_rect_color]
      • input/camera0/info [default: /sensing/camera/camera0/camera_info]
      • input/camera0/rois [default: /perception/object_recognition/detection/rois0]
      • input/camera1/image [default: /sensing/camera/camera1/image_rect_color]
      • input/camera1/info [default: /sensing/camera/camera1/camera_info]
      • input/camera1/rois [default: /perception/object_recognition/detection/rois1]
      • input/camera2/image [default: /sensing/camera/camera2/image_rect_color]
      • input/camera2/info [default: /sensing/camera/camera2/camera_info]
      • input/camera2/rois [default: /perception/object_recognition/detection/rois2]
      • input/camera3/image [default: /sensing/camera/camera3/image_rect_color]
      • input/camera3/info [default: /sensing/camera/camera3/camera_info]
      • input/camera3/rois [default: /perception/object_recognition/detection/rois3]
      • input/camera4/image [default: /sensing/camera/camera4/image_rect_color]
      • input/camera4/info [default: /sensing/camera/camera4/camera_info]
      • input/camera4/rois [default: /perception/object_recognition/detection/rois4]
      • input/camera5/image [default: /sensing/camera/camera5/image_rect_color]
      • input/camera5/info [default: /sensing/camera/camera5/camera_info]
      • input/camera5/rois [default: /perception/object_recognition/detection/rois5]
      • input/camera6/image [default: /sensing/camera/camera6/image_rect_color]
      • input/camera6/info [default: /sensing/camera/camera6/camera_info]
      • input/camera6/rois [default: /perception/object_recognition/detection/rois6]
      • input/camera7/image [default: /sensing/camera/camera7/image_rect_color]
      • input/camera7/info [default: /sensing/camera/camera7/camera_info]
      • input/camera7/rois [default: /perception/object_recognition/detection/rois7]
      • output/tracker_merged_objects [default: /perception/object_recognition/detection/objects]
      • output/tracked_objects [default: /perception/object_recognition/tracking/objects]
      • output/predicted_objects [default: /perception/object_recognition/objects]
      • near_range_camera_vru_param_path [default: $(var perception_config_dir)/object_recognition/detection/camera_vru_detection/near_range_camera_vru_detector.param.yaml]
      • object_recognition_tracking_multi_object_tracker_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/data_association_matrix.param.yaml]
      • object_recognition_tracking_multi_object_tracker_input_channels_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/input_channels.param.yaml]
      • object_recognition_tracking_multi_object_tracker_node_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/multi_object_tracker_node.param.yaml]
  • launch/object_recognition/object_recognition_camera_lidar_radar.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • lidar_detection_model [default: centerpoint]
      • use_short_range_detection [default: false]
      • lidar_short_range_detection_model_type [default: centerpoint_short_range]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • bevfusion_model_path [default: $(var data_path)/bevfusion]
      • pointpainting_model_path [default: $(var data_path)/image_projection_based_fusion]
      • camera_2d_detector/model_path [default: $(var data_path)/tensorrt_yolox/yolox-sPlus-opt-pseudoV2-T4-960x960-T4-seg16cls.onnx]
      • camera_2d_detector/label_path [default: $(var data_path)/tensorrt_yolox/label.txt]
      • camera_2d_detector/color_map_path [default: $(var data_path)/tensorrt_yolox/semseg_color_map.csv]
      • camera_2d_detector/roi_remap_path [default: $(var perception_config_dir)/traffic_light_recognition/tensorrt_yolox/roi_label_remap.csv]
      • camera_2d_detector/roi_to_semseg_remap_path [default: $(var perception_config_dir)/traffic_light_recognition/tensorrt_yolox/roi_to_semseg_label_remap.csv]
      • use_semantic_segmentation_ptv3 [default: false]
      • use_object_filter [default: true]
      • use_pointcloud_map [default: true]
      • use_validator [default: false]
      • objects_validation_method [default: obstacle_pointcloud]
      • use_low_height_cropbox [default: false]
      • use_low_intensity_cluster_filter [default: true]
      • use_image_segmentation_based_filter [default: false]
      • use_irregular_object_detector [default: true]
      • use_camera_vru_detector [default: false]
      • camera_vru_detector_rois_ids [default: [0]]
      • use_camera_only_detection [default: false]
      • use_vector_map [default: true]
      • tracker_publish_merged_objects [default: false]
      • number_of_cameras [default: 6]
      • pointcloud_container_name [default: pointcloud_container]
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • input/semantic_segmentation/pointcloud [default: /perception/segmented/pointcloud]
      • input/obstacle_segmentation/pointcloud [default: /perception/obstacle_segmentation/pointcloud]
      • input/camera0/image [default: /sensing/camera/camera0/image_rect_color]
      • input/camera0/info [default: /sensing/camera/camera0/camera_info]
      • input/camera0/rois [default: /perception/object_recognition/detection/rois0]
      • input/camera1/image [default: /sensing/camera/camera1/image_rect_color]
      • input/camera1/info [default: /sensing/camera/camera1/camera_info]
      • input/camera1/rois [default: /perception/object_recognition/detection/rois1]
      • input/camera2/image [default: /sensing/camera/camera2/image_rect_color]
      • input/camera2/info [default: /sensing/camera/camera2/camera_info]
      • input/camera2/rois [default: /perception/object_recognition/detection/rois2]
      • input/camera3/image [default: /sensing/camera/camera3/image_rect_color]
      • input/camera3/info [default: /sensing/camera/camera3/camera_info]
      • input/camera3/rois [default: /perception/object_recognition/detection/rois3]
      • input/camera4/image [default: /sensing/camera/camera4/image_rect_color]
      • input/camera4/info [default: /sensing/camera/camera4/camera_info]
      • input/camera4/rois [default: /perception/object_recognition/detection/rois4]
      • input/camera5/image [default: /sensing/camera/camera5/image_rect_color]
      • input/camera5/info [default: /sensing/camera/camera5/camera_info]
      • input/camera5/rois [default: /perception/object_recognition/detection/rois5]
      • input/camera6/image [default: /sensing/camera/camera6/image_rect_color]
      • input/camera6/info [default: /sensing/camera/camera6/camera_info]
      • input/camera6/rois [default: /perception/object_recognition/detection/rois6]
      • input/camera7/image [default: /sensing/camera/camera7/image_rect_color]
      • input/camera7/info [default: /sensing/camera/camera7/camera_info]
      • input/camera7/rois [default: /perception/object_recognition/detection/rois7]
      • input/camera8/image [default: /sensing/camera/camera8/image_rect_color]
      • input/camera8/info [default: /sensing/camera/camera8/camera_info]
      • input/camera8/rois [default: /perception/object_recognition/detection/rois8]
      • input/concatenation_info [default: /sensing/lidar/concatenated/pointcloud_info]
      • input/radar [default: /sensing/radar/detected_objects]
      • input/camera_only/objects [default: /perception/object_recognition/detection/camera_only/objects]
      • input/camera_only/channel [default: camera_streampetr]
      • image_topic_name [default: image_rect_color]
      • segmentation_pointcloud_fusion_camera_ids [default: [0,2,4]]
      • output/tracker_merged_objects [default: /perception/object_recognition/detection/objects]
      • output/tracked_objects [default: /perception/object_recognition/tracking/objects]
      • output/predicted_objects [default: /perception/object_recognition/objects]
      • sync_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/fusion_common.param.yaml]
      • roi_cluster_fusion_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/roi_cluster_fusion.param.yaml]
      • pointpainting_fusion_common_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/pointpainting_common.param.yaml]
      • irregular_object_crop_box_filter_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/crop_box_filter.param.yaml]
      • irregular_object_scan_ground_filter_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/scan_ground_filter.param.yaml]
      • irregular_object_roi_pointcloud_fusion_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/roi_pointcloud_fusion.param.yaml]
      • irregular_object_fusion_camera_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/fusion_cameras.param.yaml]
      • lidar_model_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model]
      • centerpoint_common_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model/centerpoint_common.param.yaml]
      • voxel_grid_based_euclidean_param_path [default: $(var perception_config_dir)/object_recognition/detection/clustering/voxel_grid_based_euclidean_cluster.param.yaml]
      • radar_lanelet_filtering_range_param_path [default: $(var perception_config_dir)/object_recognition/detection/object_filter/radar_lanelet_filter.param.yaml]
      • object_recognition_detection_object_sorter_radar_param_path [default: $(var perception_config_dir)/object_recognition/detection/radar_filter/detected_object_sorter.param.yaml]
      • near_range_camera_vru_param_path [default: $(var perception_config_dir)/object_recognition/detection/camera_vru_detection/near_range_camera_vru_detector.param.yaml]
      • object_recognition_detection_obstacle_pointcloud_based_validator_param_path [default: $(var perception_config_dir)/object_recognition/detection/detected_object_validation/obstacle_pointcloud_based_validator.param.yaml]
      • object_recognition_tracking_multi_object_tracker_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/data_association_matrix.param.yaml]
      • object_recognition_tracking_multi_object_tracker_input_channels_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/input_channels.param.yaml]
      • object_recognition_tracking_multi_object_tracker_node_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/multi_object_tracker_node.param.yaml]
      • euclidean_clustering/param_path [default: $(var perception_config_dir)/object_recognition/detection/euclidean_cluster/label_based_euclidean_cluster.param.yaml]
  • launch/object_recognition/object_recognition_camera_lidar_radar_serial.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • lidar_detection_model [default: centerpoint]
      • use_short_range_detection [default: false]
      • lidar_short_range_detection_model_type [default: centerpoint_short_range]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • bevfusion_model_path [default: $(var data_path)/bevfusion]
      • pointpainting_model_path [default: $(var data_path)/image_projection_based_fusion]
      • camera_2d_detector/model_path [default: $(var data_path)/tensorrt_yolox/yolox-sPlus-opt-pseudoV2-T4-960x960-T4-seg16cls.onnx]
      • camera_2d_detector/label_path [default: $(var data_path)/tensorrt_yolox/label.txt]
      • camera_2d_detector/color_map_path [default: $(var data_path)/tensorrt_yolox/semseg_color_map.csv]
      • camera_2d_detector/roi_remap_path [default: $(var perception_config_dir)/traffic_light_recognition/tensorrt_yolox/roi_label_remap.csv]
      • camera_2d_detector/roi_to_semseg_remap_path [default: $(var perception_config_dir)/traffic_light_recognition/tensorrt_yolox/roi_to_semseg_label_remap.csv]
      • use_object_filter [default: true]
      • objects_filter_method [default: lanelet_filter]
      • use_pointcloud_map [default: true]
      • use_validator [default: false]
      • objects_validation_method [default: obstacle_pointcloud]
      • use_roi_detected_object_validator [default: true]
      • use_detection_by_tracker [default: true]
      • use_low_height_cropbox [default: false]
      • use_low_intensity_cluster_filter [default: true]
      • use_image_segmentation_based_filter [default: false]
      • use_irregular_object_detector [default: true]
      • ml_camera_lidar_merger_priority_mode [default: 0]
      • use_vector_map [default: true]
      • number_of_cameras [default: 6]
      • pointcloud_container_name [default: pointcloud_container]
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • input/obstacle_segmentation/pointcloud [default: /perception/obstacle_segmentation/pointcloud]
      • input/camera0/image [default: /sensing/camera/camera0/image_rect_color]
      • input/camera0/info [default: /sensing/camera/camera0/camera_info]
      • input/camera0/rois [default: /perception/object_recognition/detection/rois0]
      • input/camera1/image [default: /sensing/camera/camera1/image_rect_color]
      • input/camera1/info [default: /sensing/camera/camera1/camera_info]
      • input/camera1/rois [default: /perception/object_recognition/detection/rois1]
      • input/camera2/image [default: /sensing/camera/camera2/image_rect_color]
      • input/camera2/info [default: /sensing/camera/camera2/camera_info]
      • input/camera2/rois [default: /perception/object_recognition/detection/rois2]
      • input/camera3/image [default: /sensing/camera/camera3/image_rect_color]
      • input/camera3/info [default: /sensing/camera/camera3/camera_info]
      • input/camera3/rois [default: /perception/object_recognition/detection/rois3]
      • input/camera4/image [default: /sensing/camera/camera4/image_rect_color]
      • input/camera4/info [default: /sensing/camera/camera4/camera_info]
      • input/camera4/rois [default: /perception/object_recognition/detection/rois4]
      • input/camera5/image [default: /sensing/camera/camera5/image_rect_color]
      • input/camera5/info [default: /sensing/camera/camera5/camera_info]
      • input/camera5/rois [default: /perception/object_recognition/detection/rois5]
      • input/camera6/image [default: /sensing/camera/camera6/image_rect_color]
      • input/camera6/info [default: /sensing/camera/camera6/camera_info]
      • input/camera6/rois [default: /perception/object_recognition/detection/rois6]
      • input/camera7/image [default: /sensing/camera/camera7/image_rect_color]
      • input/camera7/info [default: /sensing/camera/camera7/camera_info]
      • input/camera7/rois [default: /perception/object_recognition/detection/rois7]
      • input/camera8/image [default: /sensing/camera/camera8/image_rect_color]
      • input/camera8/info [default: /sensing/camera/camera8/camera_info]
      • input/camera8/rois [default: /perception/object_recognition/detection/rois8]
      • input/concatenation_info [default: /sensing/lidar/concatenated/pointcloud_info]
      • input/radar/tracked_objects [default: /sensing/radar/tracked_objects]
      • image_topic_name [default: image_rect_color]
      • segmentation_pointcloud_fusion_camera_ids [default: [0,2,4]]
      • output/tracked_objects [default: /perception/object_recognition/tracking/objects]
      • output/predicted_objects [default: /perception/object_recognition/objects]
      • sync_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/fusion_common.param.yaml]
      • roi_cluster_fusion_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/roi_cluster_fusion.param.yaml]
      • pointpainting_fusion_common_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/pointpainting_common.param.yaml]
      • irregular_object_crop_box_filter_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/crop_box_filter.param.yaml]
      • irregular_object_scan_ground_filter_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/scan_ground_filter.param.yaml]
      • irregular_object_roi_pointcloud_fusion_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/roi_pointcloud_fusion.param.yaml]
      • irregular_object_fusion_camera_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/fusion_cameras.param.yaml]
      • lidar_model_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model]
      • centerpoint_common_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model/centerpoint_common.param.yaml]
      • voxel_grid_based_euclidean_param_path [default: $(var perception_config_dir)/object_recognition/detection/clustering/voxel_grid_based_euclidean_cluster.param.yaml]
      • detection_by_tracker_param_path [default: $(var perception_config_dir)/object_recognition/detection/detection_by_tracker/detection_by_tracker.param.yaml]
      • object_recognition_detection_obstacle_pointcloud_based_validator_param_path [default: $(var perception_config_dir)/object_recognition/detection/detected_object_validation/obstacle_pointcloud_based_validator.param.yaml]
      • object_recognition_detection_object_merger_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/detection/object_merger/data_association_matrix.param.yaml]
      • object_recognition_detection_object_merger_distance_threshold_list_path [default: $(var perception_config_dir)/object_recognition/detection/object_merger/overlapped_judge.param.yaml]
      • ml_camera_lidar_object_association_merger_param_path [default: $(var perception_config_dir)/object_recognition/detection/object_merger/ml_camera_lidar_object_association_merger.param.yaml]
      • roi_detected_object_fusion_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/roi_detected_object_fusion.param.yaml]
      • object_recognition_tracking_multi_object_tracker_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/data_association_matrix.param.yaml]
      • object_recognition_tracking_multi_object_tracker_input_channels_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/input_channels.param.yaml]
      • object_recognition_tracking_multi_object_tracker_node_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/multi_object_tracker_node.param.yaml]
      • object_recognition_tracking_radar_tracked_object_sorter_param_path [default: $(var perception_config_dir)/object_recognition/tracking/radar_object_tracker/tracked_object_sorter.param.yaml]
      • object_recognition_tracking_radar_tracked_object_lanelet_filter_param_path [default: $(var perception_config_dir)/object_recognition/tracking/radar_object_tracker/tracked_object_lanelet_filter.param.yaml]
      • object_recognition_tracking_object_merger_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/tracking/tracking_object_merger/data_association_matrix.param.yaml]
      • object_recognition_tracking_object_merger_node_param_path [default: $(var perception_config_dir)/object_recognition/tracking/tracking_object_merger/decorative_tracker_merger.param.yaml]
      • euclidean_clustering/shape_policy [default: 0]
  • launch/object_recognition/object_recognition_lidar.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • lidar_detection_model [default: centerpoint]
      • use_short_range_detection [default: false]
      • lidar_short_range_detection_model_type [default: centerpoint_short_range]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • bevfusion_model_path [default: $(var data_path)/bevfusion]
      • use_pointcloud_map [default: true]
      • use_validator [default: false]
      • objects_validation_method [default: obstacle_pointcloud]
      • use_object_filter [default: true]
      • objects_filter_method [default: lanelet_filter]
      • use_low_height_cropbox [default: false]
      • use_vector_map [default: true]
      • tracker_publish_merged_objects [default: false]
      • pointcloud_container_name [default: pointcloud_container]
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • input/obstacle_segmentation/pointcloud [default: /perception/obstacle_segmentation/pointcloud]
      • output/tracker_merged_objects [default: /perception/object_recognition/detection/objects]
      • output/tracked_objects [default: /perception/object_recognition/tracking/objects]
      • output/predicted_objects [default: /perception/object_recognition/objects]
      • lidar_model_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model]
      • centerpoint_common_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model/centerpoint_common.param.yaml]
      • voxel_grid_based_euclidean_param_path [default: $(var perception_config_dir)/object_recognition/detection/clustering/voxel_grid_based_euclidean_cluster.param.yaml]
      • object_recognition_detection_obstacle_pointcloud_based_validator_param_path [default: $(var perception_config_dir)/object_recognition/detection/detected_object_validation/obstacle_pointcloud_based_validator.param.yaml]
      • object_recognition_tracking_multi_object_tracker_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/data_association_matrix.param.yaml]
      • object_recognition_tracking_multi_object_tracker_input_channels_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/input_channels.param.yaml]
      • object_recognition_tracking_multi_object_tracker_node_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/multi_object_tracker_node.param.yaml]
  • launch/object_recognition/object_recognition_lidar_ptv3.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • lidar_detection_model [default: centerpoint]
      • use_short_range_detection [default: false]
      • lidar_short_range_detection_model_type [default: centerpoint_short_range]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • bevfusion_model_path [default: $(var data_path)/bevfusion]
      • use_vector_map [default: true]
      • tracker_publish_merged_objects [default: false]
      • pointcloud_container_name [default: pointcloud_container]
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • input/semantic_segmentation/pointcloud [default: /perception/segmented/pointcloud]
      • output/tracker_merged_objects [default: /perception/object_recognition/detection/objects]
      • output/tracked_objects [default: /perception/object_recognition/tracking/objects]
      • output/predicted_objects [default: /perception/object_recognition/objects]
      • lidar_model_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model]
      • centerpoint_common_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model/centerpoint_common.param.yaml]
      • object_recognition_tracking_multi_object_tracker_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/data_association_matrix.param.yaml]
      • object_recognition_tracking_multi_object_tracker_input_channels_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/input_channels.param.yaml]
      • object_recognition_tracking_multi_object_tracker_node_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/multi_object_tracker_node.param.yaml]
      • euclidean_clustering/param_path [default: $(var perception_config_dir)/object_recognition/detection/euclidean_cluster/label_based_euclidean_cluster.param.yaml]
  • launch/object_recognition/object_recognition_lidar_radar.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • lidar_detection_model [default: centerpoint]
      • use_short_range_detection [default: false]
      • lidar_short_range_detection_model_type [default: centerpoint_short_range]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • bevfusion_model_path [default: $(var data_path)/bevfusion]
      • use_low_height_cropbox [default: false]
      • use_vector_map [default: true]
      • tracker_publish_merged_objects [default: false]
      • pointcloud_container_name [default: pointcloud_container]
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • input/semantic_segmentation/pointcloud [default: /perception/segmented/pointcloud]
      • input/radar [default: /sensing/radar/detected_objects]
      • output/tracker_merged_objects [default: /perception/object_recognition/detection/objects]
      • output/tracked_objects [default: /perception/object_recognition/tracking/objects]
      • output/predicted_objects [default: /perception/object_recognition/objects]
      • lidar_model_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model]
      • centerpoint_common_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model/centerpoint_common.param.yaml]
      • voxel_grid_based_euclidean_param_path [default: $(var perception_config_dir)/object_recognition/detection/clustering/voxel_grid_based_euclidean_cluster.param.yaml]
      • radar_lanelet_filtering_range_param_path [default: $(var perception_config_dir)/object_recognition/detection/object_filter/radar_lanelet_filter.param.yaml]
      • object_recognition_detection_object_sorter_radar_param_path [default: $(var perception_config_dir)/object_recognition/detection/radar_filter/detected_object_sorter.param.yaml]
      • object_recognition_tracking_multi_object_tracker_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/data_association_matrix.param.yaml]
      • object_recognition_tracking_multi_object_tracker_input_channels_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/input_channels.param.yaml]
      • object_recognition_tracking_multi_object_tracker_node_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/multi_object_tracker_node.param.yaml]
      • euclidean_clustering/param_path [default: $(var perception_config_dir)/object_recognition/detection/euclidean_cluster/label_based_euclidean_cluster.param.yaml]
  • launch/object_recognition/prediction/prediction.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • use_vector_map [default: false]
      • prediction_model_type [default: map_based]
      • input/objects [default: /perception/object_recognition/tracking/objects]
      • output/objects [default: /perception/object_recognition/objects]
      • object_recognition_prediction_map_based_prediction_param_path [default: $(var perception_config_dir)/object_recognition/prediction/map_based_prediction.param.yaml]
  • launch/obstacle_segmentation/ground_segmentation_additional_lidars.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • target_container
      • lidar_name
      • base_frame [default: base_link]
      • use_intra_process [default: true]
      • input/main/pointcloud
      • input/lidar/pointcloud [default: /sensing/lidar/$(var lidar_name)/pointcloud]
      • output/pointcloud
      • output/pointcloud_info [default: $(var output/pointcloud)_info]
      • crop_box_margin_max_z [default: 2.5]
      • crop_box_margin_min_z [default: -2.5]
      • crop_box_filter_node_param_path [default: $(var perception_config_dir)/obstacle_segmentation/ground_segmentation/crop_box_filter.param.yaml]
      • scan_ground_filter_node_param_path [default: $(var perception_config_dir)/obstacle_segmentation/ground_segmentation/scan_ground_filter.param.yaml]
  • launch/obstacle_segmentation/ground_segmentation_ml.launch.xml
      • pointcloud_container_name [default: pointcloud_container]
      • use_pointcloud_container [default: true]
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • output/pointcloud [default: /perception/obstacle_segmentation/pointcloud]
      • output/pointcloud/segmentation [default: /perception/segmented/pointcloud]
      • output/pointcloud/visualization [default: /perception/segmented/pointcloud/visualization]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • model_variant [default: ptv3]
      • model_path [default: $(var data_path)/ptv3]
  • launch/obstacle_segmentation/ground_segmentation_ransac.launch.xml
      • target_container
      • base_frame [default: base_link]
      • use_intra_process [default: true]
      • input_topics
      • input/main/pointcloud
      • output/pointcloud
      • output/pointcloud_info [default: $(var output/pointcloud)_info]
      • detection_area_crop_box_filter_node_param_path
      • ransac_ground_filter_node_param_path [default: $(find-pkg-share autoware_ground_segmentation)/config/ransac_ground_filter.param.yaml]
  • launch/obstacle_segmentation/ground_segmentation_rule_based.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • target_container
      • base_frame [default: base_link]
      • use_intra_process [default: true]
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • output/pointcloud [default: /perception/obstacle_segmentation/pointcloud]
      • crop_box_margin_max_z [default: 2.5]
      • crop_box_margin_min_z [default: -2.5]
      • crop_box_filter_node_param_path [default: $(var perception_config_dir)/obstacle_segmentation/ground_segmentation/crop_box_filter.param.yaml]
      • scan_ground_filter_node_param_path [default: $(var perception_config_dir)/obstacle_segmentation/ground_segmentation/scan_ground_filter.param.yaml]
  • launch/obstacle_segmentation/ground_segmentation_rule_based_cuda.launch.xml
      • target_container
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • output/pointcloud [default: /perception/obstacle_segmentation/pointcloud]
      • output/ground_pointcloud [default: /perception/obstacle_segmentation/ground_pointcloud]
      • cuda_scan_ground_segmentation_filter_node_param_path [default: $(find-pkg-share autoware_ground_segmentation_cuda)/config/cuda_scan_ground_segmentation_filter.param.yaml]
  • launch/occupancy_grid_map/probabilistic_occupancy_grid_map.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • input/obstacle_pointcloud [default: concatenated/pointcloud]
      • input/raw_pointcloud [default: no_ground/oneshot/pointcloud]
      • output [default: /perception/occupancy_grid_map/map]
      • use_intra_process [default: false]
      • use_multithread [default: false]
      • pointcloud_container_name [default: pointcloud_container]
      • occupancy_grid_map_method
      • occupancy_grid_map_param_path [default: $(var perception_config_dir)/occupancy_grid_map/$(var occupancy_grid_map_method).param.yaml]
      • occupancy_grid_map_updater
      • occupancy_grid_map_updater_param_path [default: $(var perception_config_dir)/occupancy_grid_map/$(var occupancy_grid_map_updater)_updater.param.yaml]
      • input_obstacle_pointcloud [default: false]
      • input_obstacle_and_raw_pointcloud [default: true]
      • use_pointcloud_container [default: true]
  • launch/traffic_light_recognition/traffic_light_detector_fine.launch.xml
    • Fine-detection chain for one traffic light camera: map based detector proposes rough ROIs, the fine detector refines them, classifiers label them.
      • perception_config_pkg [default: autoware_perception_config]
      • camera_namespace
      • enable_image_decompressor [default: true]
      • input/vector_map [default: /map/vector_map]
      • input/route [default: /planning/mission_planning/route]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • map_based_detector_param_path [default: $(find-pkg-share autoware_traffic_light_map_based_detector)/config/traffic_light_map_based_detector.param.yaml]
      • fine_detector_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_fine_detector/traffic_light_fine_detector.param.yaml]
      • car_classifier_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_classifier/car_traffic_light_classifier.param.yaml]
      • pedestrian_classifier_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_classifier/pedestrian_traffic_light_classifier.param.yaml]
      • lamp_recognizer_ml_param_path [default: $(var data_path)/traffic_light_classifier/lamp_recognizer_ml.param.yaml]
      • roi_visualizer_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_visualization/traffic_light_roi_visualizer.param.yaml]
      • fine_detector_model_path [default: $(var data_path)/traffic_light_fine_detector/tlr_car_ped_yolox_s_batch_6.onnx]
      • fine_detector_label_path [default: $(var data_path)/traffic_light_fine_detector/tlr_labels.txt]
      • car_classifier_model_path [default: $(var data_path)/traffic_light_classifier/traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • car_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels.txt]
      • pedestrian_classifier_model_path [default: $(var data_path)/traffic_light_classifier/ped_traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • pedestrian_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels_ped.txt]
      • car_classifier_type [default: 1]
      • pedestrian_classifier_type [default: 1]
  • launch/traffic_light_recognition/traffic_light_detector_rough.launch.xml
    • Rough-detection chain for one traffic light camera: map based detector ROIs feed the classifiers directly, without a high accuracy detector.
      • perception_config_pkg [default: autoware_perception_config]
      • camera_namespace
      • enable_image_decompressor [default: true]
      • input/vector_map [default: /map/vector_map]
      • input/route [default: /planning/mission_planning/route]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • map_based_detector_param_path [default: $(find-pkg-share autoware_traffic_light_map_based_detector)/config/traffic_light_map_based_detector.param.yaml]
      • car_classifier_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_classifier/car_traffic_light_classifier.param.yaml]
      • pedestrian_classifier_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_classifier/pedestrian_traffic_light_classifier.param.yaml]
      • lamp_recognizer_ml_param_path [default: $(var data_path)/traffic_light_classifier/lamp_recognizer_ml.param.yaml]
      • roi_visualizer_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_visualization/traffic_light_roi_visualizer.param.yaml]
      • car_classifier_model_path [default: $(var data_path)/traffic_light_classifier/traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • car_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels.txt]
      • pedestrian_classifier_model_path [default: $(var data_path)/traffic_light_classifier/ped_traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • pedestrian_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels_ped.txt]
      • car_classifier_type [default: 1]
      • pedestrian_classifier_type [default: 1]
  • launch/traffic_light_recognition/traffic_light_detector_whole_image.launch.xml
    • Whole-image detection chain for one traffic light camera: yolox detects over the full image, the selector matches detections to map based ROIs, classifiers label them.
      • perception_config_pkg [default: autoware_perception_config]
      • camera_namespace
      • enable_image_decompressor [default: true]
      • input/vector_map [default: /map/vector_map]
      • input/route [default: /planning/mission_planning/route]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • map_based_detector_param_path [default: $(find-pkg-share autoware_traffic_light_map_based_detector)/config/traffic_light_map_based_detector.param.yaml]
      • whole_image_detector_param_path [default: $(var perception_config_dir)/traffic_light_recognition/tensorrt_yolox/yolox_traffic_light_detector.param.yaml]
      • car_classifier_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_classifier/car_traffic_light_classifier.param.yaml]
      • pedestrian_classifier_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_classifier/pedestrian_traffic_light_classifier.param.yaml]
      • lamp_recognizer_ml_param_path [default: $(var data_path)/traffic_light_classifier/lamp_recognizer_ml.param.yaml]
      • roi_visualizer_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_visualization/traffic_light_roi_visualizer.param.yaml]
      • whole_image_detector_model_path [default: $(var data_path)/tensorrt_yolox/yolox_s_car_ped_tl_detector_960_960_batch_1.onnx]
      • whole_image_detector_label_path [default: $(var data_path)/tensorrt_yolox/car_ped_tl_detector_labels.txt]
      • car_classifier_model_path [default: $(var data_path)/traffic_light_classifier/traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • car_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels.txt]
      • pedestrian_classifier_model_path [default: $(var data_path)/traffic_light_classifier/ped_traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • pedestrian_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels_ped.txt]
      • car_classifier_type [default: 1]
      • pedestrian_classifier_type [default: 1]
  • launch/traffic_light_recognition/traffic_light_fusion.launch.xml
    • Post-process shared by all traffic light cameras: multi camera fusion, V2X arbitration, pedestrian signal estimation and map visualization.
      • perception_config_pkg [default: autoware_perception_config]
      • camera_namespaces [default: [camera6, camera7]]
      • input/vector_map [default: /map/vector_map]
      • external/traffic_signals [default: /perception/traffic_light_recognition/external/traffic_signals]
      • internal/traffic_signals [default: /perception/traffic_light_recognition/internal/traffic_signals]
      • judged/traffic_signals [default: /perception/traffic_light_recognition/judged/traffic_signals]
      • output/traffic_signals [default: /perception/traffic_light_recognition/traffic_signals]
      • multi_camera_fusion_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_multi_camera_fusion/traffic_light_multi_camera_fusion.param.yaml]
      • arbiter_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_arbiter/traffic_light_arbiter.param.yaml]
      • crosswalk_estimator_param_path [default: $(var perception_config_dir)/traffic_light_recognition/crosswalk_traffic_light_estimator/crosswalk_traffic_light_estimator.param.yaml]
  • launch/traffic_light_recognition/traffic_light_occlusion_predictor.launch.xml
    • Occlusion check for one traffic light camera: merges the car and pedestrian signals into the per-camera output, masking signals hidden from the camera.
      • perception_config_pkg [default: autoware_perception_config]
      • camera_namespace
      • input/cloud [default: /sensing/lidar/top/pointcloud_raw_ex]
      • input/vector_map [default: /map/vector_map]
      • occlusion_predictor_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_occlusion_predictor/traffic_light_occlusion_predictor.param.yaml]
  • launch/traffic_light_recognition/traffic_light_recognition_fine.launch.xml
    • Traffic light recognition, fine detection mode: per-camera fine detection with occlusion check, then shared fusion. Two cameras as the representative set; add or remove per-camera blocks to match the vehicle.
      • perception_config_pkg [default: autoware_perception_config]
      • input/vector_map [default: /map/vector_map]
      • input/route [default: /planning/mission_planning/route]
      • input/pointcloud [default: /sensing/lidar/top/pointcloud_raw_ex]
      • external/traffic_signals [default: /perception/traffic_light_recognition/external/traffic_signals]
      • output/traffic_signals [default: /perception/traffic_light_recognition/traffic_signals]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • fine_detector_model_path [default: $(var data_path)/traffic_light_fine_detector/tlr_car_ped_yolox_s_batch_6.onnx]
      • fine_detector_label_path [default: $(var data_path)/traffic_light_fine_detector/tlr_labels.txt]
      • lamp_recognizer_ml_param_path [default: $(var data_path)/traffic_light_classifier/lamp_recognizer_ml.param.yaml]
      • car_classifier_model_path [default: $(var data_path)/traffic_light_classifier/traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • car_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels.txt]
      • pedestrian_classifier_model_path [default: $(var data_path)/traffic_light_classifier/ped_traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • pedestrian_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels_ped.txt]
      • car_classifier_type [default: 1]
      • pedestrian_classifier_type [default: 1]
      • base_map_based_detector_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_map_based_detector/base_traffic_light_map_based_detector.param.yaml]
      • extra_map_based_detector_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_map_based_detector/extra_traffic_light_map_based_detector.param.yaml]
  • launch/traffic_light_recognition/traffic_light_recognition_merge.launch.xml
    • Traffic light recognition, merge mode: detection and classification run on other ECUs; this side runs the per-camera occlusion check and the shared fusion. Two cameras as the representative set; add or remove per-camera blocks to match the vehicle.
      • input/vector_map [default: /map/vector_map]
      • input/pointcloud [default: /sensing/lidar/top/pointcloud_raw_ex]
      • external/traffic_signals [default: /perception/traffic_light_recognition/external/traffic_signals]
      • output/traffic_signals [default: /perception/traffic_light_recognition/traffic_signals]
  • launch/traffic_light_recognition/traffic_light_recognition_rough.launch.xml
    • Traffic light recognition, rough detection mode: per-camera map based detection with occlusion check, then shared fusion. One camera as the representative set; add or remove per-camera blocks to match the vehicle.
      • perception_config_pkg [default: autoware_perception_config]
      • input/vector_map [default: /map/vector_map]
      • input/route [default: /planning/mission_planning/route]
      • input/pointcloud [default: /sensing/lidar/top/pointcloud_raw_ex]
      • external/traffic_signals [default: /perception/traffic_light_recognition/external/traffic_signals]
      • output/traffic_signals [default: /perception/traffic_light_recognition/traffic_signals]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • lamp_recognizer_ml_param_path [default: $(var data_path)/traffic_light_classifier/lamp_recognizer_ml.param.yaml]
      • car_classifier_model_path [default: $(var data_path)/traffic_light_classifier/traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • car_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels.txt]
      • pedestrian_classifier_model_path [default: $(var data_path)/traffic_light_classifier/ped_traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • pedestrian_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels_ped.txt]
      • car_classifier_type [default: 1]
      • pedestrian_classifier_type [default: 1]
      • base_map_based_detector_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_map_based_detector/base_traffic_light_map_based_detector.param.yaml]
  • launch/traffic_light_recognition/traffic_light_recognition_whole_image.launch.xml
    • Traffic light recognition, whole-image detection mode: per-camera yolox detection, then shared fusion. Two cameras as the representative set; add or remove per-camera blocks to match the vehicle.
      • perception_config_pkg [default: autoware_perception_config]
      • input/vector_map [default: /map/vector_map]
      • input/route [default: /planning/mission_planning/route]
      • external/traffic_signals [default: /perception/traffic_light_recognition/external/traffic_signals]
      • output/traffic_signals [default: /perception/traffic_light_recognition/traffic_signals]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • whole_image_detector_model_path [default: $(var data_path)/tensorrt_yolox/yolox_s_car_ped_tl_detector_960_960_batch_1.onnx]
      • whole_image_detector_label_path [default: $(var data_path)/tensorrt_yolox/car_ped_tl_detector_labels.txt]
      • lamp_recognizer_ml_param_path [default: $(var data_path)/traffic_light_classifier/lamp_recognizer_ml.param.yaml]
      • car_classifier_model_path [default: $(var data_path)/traffic_light_classifier/traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • car_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels.txt]
      • pedestrian_classifier_model_path [default: $(var data_path)/traffic_light_classifier/ped_traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • pedestrian_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels_ped.txt]
      • car_classifier_type [default: 1]
      • pedestrian_classifier_type [default: 1]
      • base_map_based_detector_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_map_based_detector/base_traffic_light_map_based_detector.param.yaml]
      • extra_map_based_detector_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_map_based_detector/extra_traffic_light_map_based_detector.param.yaml]

Messages

No message files found.

Services

No service files found

Plugins

No plugins found.

Recent questions tagged autoware_perception_launch at Robotics Stack Exchange

No version for distro melodic showing github. Known supported distros are highlighted in the buttons above.
Package symbol

autoware_perception_launch package from autoware_launch repo

autoware_launch autoware_sample_designs autoware_control_config autoware_control_launch autoware_localization_config autoware_localization_launch autoware_map_config autoware_map_launch autoware_perception_config autoware_perception_launch autoware_planning_config autoware_planning_launch autoware_sensing_launch autoware_sensor_driver_launch awsim_labs_sensor_kit_description awsim_labs_sensor_kit_launch awsim_sensor_kit_description awsim_sensor_kit_launch carla_sensor_kit_description carla_sensor_kit_launch sample_sensor_kit_description sample_sensor_kit_launch single_lidar_sensor_kit_description single_lidar_sensor_kit_launch tier4_autoware_api_launch tier4_simulator_launch tier4_system_launch tier4_vehicle_launch awsim_labs_vehicle_description awsim_labs_vehicle_launch sample_vehicle_description sample_vehicle_launch

ROS Distro
github

API Docs
Browse Code
Wiki

Package Summary

Version 0.53.0
License Apache License 2.0
Build type AMENT_CMAKE
Use RECOMMENDED

Repository Summary

Description
Checkout URI https://github.com/autowarefoundation/autoware_launch.git
VCS Type git
VCS Version main
Last Updated 2026-10-26
Dev Status UNKNOWN
Released UNRELEASED
Contributing Help Wanted (-)
Good First Issues (-)
Pull Requests to Review (-)

Package Description

The autoware_perception_launch package

Maintainers

  • Yukihiro Saito
  • Dai Nguyen
  • Taekjin Lee
  • Yoshi Ri
  • Masato Saeki

Authors

No additional authors.
autoware_universe_launch/autoware_perception_launch/README.md

autoware_perception_launch

Shared perception launch library. The system component launchers live in the product layer (autoware_launch/launch/components/component_perception.launch.xml for the lidar pipeline and component_traffic_light.launch.xml for traffic light recognition); this package hosts the perception feature launchers they combine:

launch/
├── common/                        # shared preprocess feature launchers (pointcloud downsample)
├── object_recognition/
│   ├── object_recognition_<mode>.launch.xml  # detection + tracking + prediction per sensor mode
│   ├── detection/
│   │   ├── detector/              # detectors
│   │   ├── filter/                # filters
│   │   └── merger/                # detection-side serial merger (used by the camera_lidar_radar_serial launcher)
│   └── prediction/
├── obstacle_segmentation/
├── occupancy_grid_map/
└── traffic_light_recognition/
design/module/                     # Autoware System Designer modules; same subtree as launch/
├── object_recognition/
├── obstacle_segmentation/
├── occupancy_grid_map/
└── traffic_light_recognition/

Launch files

  • object_recognition/object_recognition_<mode>.launch.xml — one feature launcher per sensor mode (camera, lidar, lidar_ptv3, lidar_radar, camera_lidar_radar, camera_lidar_radar_serial); hosts the detection stage inline together with the tracker and prediction, and pushes their module namespaces. Camera and radar input topics default to their absolute sensing paths (/sensing/camera/cameraN/..., /sensing/radar/detected_objects), so callers override only what differs. Most modes wire the multi-channel tracker to the detector outputs; camera_lidar_radar_serial instead merges the detectors in the detection stage (optionally regenerating detections through detection-by-tracker), tracks the single merged detected_objects channel, and joins radar tracked objects at the decorative tracker merger. camera runs camera-only detectors (BEV and camera VRU).
  • The clustering backend is file selection, not a flag: lidar runs rule-based euclidean clustering off the obstacle segmentation cloud, while lidar_ptv3 and lidar_radar cluster the PTv3 semantic-segmentation cloud and drop the pointcloud map filter and the object validator that only the rule-based pipeline uses. Both feed their cluster objects to the tracker on the lidar_clustering channel.
  • object_recognition/detection/{detector,filter}/ — single-purpose launchers wired together by the object-recognition launchers. merger/camera_lidar_merger.launch.xml chains the detection-side object mergers for the camera_lidar_radar_serial launcher.
  • obstacle_segmentation/ground_segmentation_<backend>.launch.xml — ground removal backends (rule_based, rule_based_cuda, ml); backend selection is file selection. ground_segmentation_{additional_lidars,ransac}.launch.xml fold extra lidars into the obstacle stream for kits that carry them.
  • occupancy_grid_map/, common/, traffic_light_recognition/ — feature launchers included directly by the system components.

Namespace and container addressing

Callers own the namespaces: the system component pushes perception and object_recognition; object_recognition_<mode>.launch.xml pushes detection/tracking/prediction. The object-recognition launchers address their internal topics through the fixed /perception/object_recognition namespace. Composable nodes load into the shared pointcloud container, whose name is exposed as pointcloud_container_name and aliased to the node/pointcloud_container argument of the detector launchers.

Config

Parameter files are resolved from a config/ tree addressed through the perception_config_pkg argument, which defaults to autoware_perception_config. A product can relocate the whole tree by setting perception_config_pkg once at its entry point — the launch configuration propagates through every include — but it must then ship the same config/ layout. Products that differ in only a few files instead override the corresponding *_param_path arguments, which every feature launcher declares with a default:

<include file="$(find-pkg-share autoware_perception_launch)/launch/object_recognition/object_recognition_lidar.launch.xml">
  <arg name="perception_config_pkg" value="my_perception_config"/>
  <arg name="FOO_NODE_param_path" value="..."/>
</include>

component_perception.launch.xml, component_traffic_light.launch.xml and planning_simulator/perception_simulation.launch.xml each declare perception_config_pkg, so a product switches the tree at the component it launches.

The launch/ tree is the interface downstream distributions adopt as-is; only package.xml, CMakeLists.txt, README.md, and changelogs stay package-local. Keep changes to this tree folder-merge compatible.

Package Dependencies

Please see <exec_depend> in package.xml.

CHANGELOG

Changelog for package autoware_perception_launch

0.53.0 (2026-10-13)

  • Merge remote-tracking branch 'origin/main' into tmp/bot/bump_version_base

  • feat(autoware_launch): run crop_box_filter standalone when agnocast is enabled (#1990)

    * feat(autoware_launch): run crop_box_filter standalone when agnocast is enabled autoware_universe#13384 moves CropBoxFilterComponent to agnocast_wrapper::Node on an agnocast-only executor, which the component containers do not set up. Start each of the eight crop_box_filter nodes as a standalone process when ENABLE_AGNOCAST=1 and keep loading them into their container otherwise. The node names, namespaces, remaps and parameters are the same in both branches.

    * refactor(autoware_launch): launch crop_box_filter through the autoware_node action ---------

  • feat(autoware_perception_launch): run scan_ground_filter standalone when agnocast is enabled (#1991)

    * feat(autoware_perception_launch): run scan_ground_filter standalone when agnocast is enabled autoware_universe#13385 moves ScanGroundFilterComponent to agnocast_wrapper::Node on an agnocast-only executor, which the component containers do not set up. Start each of the three scan_ground_filter nodes as a standalone process when ENABLE_AGNOCAST=1 and keep loading them into their container otherwise. The node names, namespaces, remaps and parameters are the same in both branches.

    * refactor(autoware_perception_launch): launch scan_ground_filter through the autoware_node action ---------

  • feat(autoware_launch): run voxel_grid_downsample_filter standalone when agnocast is enabled (#1992)

    • feat(autoware_launch): run voxel_grid_downsample_filter standalone when agnocast is enabled
    • style(pre-commit): autofix

    * refactor(autoware_launch): launch voxel_grid_downsample_filter through the autoware_node action ---------Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>

  • feat(autoware_sample_designs): compose the AD API, system and vehicle components from design modules (#1976)

    * fix(control, planning): connect the operation mode state and velocity limit inputs Wires collision_detector's operation_mode_state and remaining_distance_time_calculator's planning_velocity through the module designs; their node designs now declare the fixed topic names as remap targets.

    * fix(planning): connect the internal velocity limit path to the selector Forwards motion_velocity_planner's velocity_limit and clear_velocity_limit up to ScenarioPlanning and feeds them into external_velocity_limit_selector, whose internal inputs were unconnected.

    * fix(control, planning): connect the remaining module inputs to their sources Wires the operation mode transition manager's gate and selector inputs, the stop mode operator's vehicle status, the manual lane change handler's route availability and preferred primitive service, the parking costmap, and the route marker. Exposes the vehicle control mode service, the preferred lane service and the system hazard lights command as component interfaces.

    • fix(planning): follow the hazard lights selector input rename
    • fix(planning): add steering_status subscriber to various planning modules
    • feat(vehicle): add vehicle wrapper and launch file for vehicle interface
    • style(pre-commit): autofix
    • fix(planning): add planning factors publishers and connections for various modules
    • refactor(autoware_sample_designs): compose the AD API from design modules
    • Refactor ADAPI module structure and enhance planning factors
    • Updated AdapiRvizHelpers.module.yaml to include helper prefixes for instances and connections.
    • Modified AutowareSample.system.yaml to reorganize ADAPI components, introducing default_adapi for better structure.
    • Expanded planning factors in Planning.module.yaml, LaneDriving.module.yaml, ScenarioPlanning.module.yaml, and associated modules to include additional factors for improved behavior and motion planning.
    • Enhanced connections in behavior planning and motion planning modules to accommodate new planning factors.
    • Ensured consistency across scenario planning modules by aligning

File truncated at 100 lines see the full file

Package Dependencies

Deps Name
ament_cmake_auto
autoware_cmake
autoware_agnocast_wrapper
autoware_bevfusion
autoware_bytetrack
autoware_compare_map_segmentation
autoware_crosswalk_traffic_light_estimator
autoware_cuda_pointcloud_preprocessor
autoware_detected_object_feature_remover
autoware_detected_object_validation
autoware_detection_by_tracker
autoware_elevation_map_loader
autoware_euclidean_cluster
autoware_ground_segmentation
autoware_ground_segmentation_cuda
autoware_image_object_locator
autoware_image_projection_based_fusion
autoware_image_transport_decompressor
autoware_lidar_apollo_instance_segmentation
autoware_lidar_centerpoint
autoware_lidar_transfusion
autoware_map_based_prediction
autoware_multi_object_tracker
autoware_object_merger
autoware_object_sorter
autoware_occupancy_grid_map_outlier_filter
autoware_perception_config
autoware_pointcloud_preprocessor
autoware_probabilistic_occupancy_grid_map
autoware_ptv3
autoware_raindrop_cluster_filter
autoware_shape_estimation
autoware_simpl_prediction
autoware_simple_object_merger
autoware_tensorrt_bevdet
autoware_tensorrt_yolox
autoware_tracking_object_merger
autoware_traffic_light_arbiter
autoware_traffic_light_category_merger
autoware_traffic_light_classifier
autoware_traffic_light_fine_detector
autoware_traffic_light_map_based_detector
autoware_traffic_light_multi_camera_fusion
autoware_traffic_light_occlusion_predictor
autoware_traffic_light_selector
autoware_traffic_light_visualization
rclcpp_components
topic_tools
ament_lint_auto
autoware_lint_common

System Dependencies

No direct system dependencies.

Dependant Packages

Name Deps
autoware_launch
reaction_analyzer

Launch files

  • launch/common/pointcloud_downsample.launch.xml
      • pointcloud_container_name
      • cuda_pointcloud_preprocessing [default: false]
      • input/pointcloud
      • output/pointcloud
      • common_downsample_voxel_size_x [default: 0.05]
      • common_downsample_voxel_size_y [default: 0.05]
      • common_downsample_voxel_size_z [default: 0.05]
  • launch/object_recognition/detection/detector/camera_2d_detector.launch.xml
      • image_raw0 [default: /sensing/camera/camera0/image_raw]
      • image_raw1 [default: /sensing/camera/camera1/image_raw]
      • image_raw2 [default: /sensing/camera/camera2/image_raw]
      • image_raw3 [default: /sensing/camera/camera3/image_raw]
      • image_raw4 [default: /sensing/camera/camera4/image_raw]
      • image_raw5 [default: /sensing/camera/camera5/image_raw]
      • image_raw6 [default: /sensing/camera/camera6/image_raw]
      • image_raw7 [default: /sensing/camera/camera7/image_raw]
      • image_raw8 [default: /sensing/camera/camera8/image_raw]
      • image_raw9 [default: /sensing/camera/camera9/image_raw]
      • image_number [default: 1]
      • camera_index [default: 0]
      • use_bytetrack [default: true]
      • enable_visualizer [default: false]
      • camera_2d_detector/model_path
      • camera_2d_detector/label_path
      • camera_2d_detector/color_map_path
      • camera_2d_detector/roi_remap_path
      • camera_2d_detector/roi_to_semseg_remap_path
      • tensorrt_yolox_ns [default: ]
  • launch/object_recognition/detection/detector/camera_bev_detector.launch.xml
      • input/camera0/image
      • input/camera0/info
      • input/camera1/image
      • input/camera1/info
      • input/camera2/image
      • input/camera2/info
      • input/camera3/image
      • input/camera3/info
      • input/camera4/image
      • input/camera4/info
      • input/camera5/image
      • input/camera5/info
      • output/objects
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • bevdet_model_name [default: bevdet_one_lt_d]
      • bevdet_model_path [default: $(var data_path)/tensorrt_bevdet]
  • launch/object_recognition/detection/detector/camera_lidar_detector.launch.xml
      • ns
      • lidar_detection_model_type
      • lidar_detection_model_name
      • use_low_intensity_cluster_filter
      • use_image_segmentation_based_filter
      • number_of_cameras
      • input/camera0/image
      • input/camera0/info
      • input/camera0/rois
      • input/camera1/image
      • input/camera1/info
      • input/camera1/rois
      • input/camera2/image
      • input/camera2/info
      • input/camera2/rois
      • input/camera3/image
      • input/camera3/info
      • input/camera3/rois
      • input/camera4/image
      • input/camera4/info
      • input/camera4/rois
      • input/camera5/image
      • input/camera5/info
      • input/camera5/rois
      • input/camera6/image
      • input/camera6/info
      • input/camera6/rois
      • input/camera7/image
      • input/camera7/info
      • input/camera7/rois
      • input/camera8/image
      • input/camera8/info
      • input/camera8/rois
      • input/concatenation_info
      • segmentation_pointcloud_fusion_camera_ids
      • image_topic_name
      • sync_param_path
      • voxel_grid_based_euclidean_param_path
      • lidar_model_param_path
      • pointpainting_model_path [default: $(var data_path)/image_projection_based_fusion]
      • pointpainting_fusion_common_param_path
      • roi_cluster_fusion_param_path
      • node/pointcloud_container
      • input/pointcloud
      • input/pointcloud_map/pointcloud
      • input/obstacle_segmentation/pointcloud
      • output/ml_detector/objects
      • output/rule_detector/objects
      • camera_2d_detector/model_path
      • camera_2d_detector/label_path
      • camera_2d_detector/color_map_path
      • camera_2d_detector/roi_remap_path
      • camera_2d_detector/roi_to_semseg_remap_path
      • enable_2d_detection [default: false]
  • launch/object_recognition/detection/detector/camera_lidar_irregular_object_detector.launch.xml
      • pipeline_ns
      • node/pointcloud_container
      • output/objects [default: objects]
      • input/concatenation_info
      • input/pointcloud
      • irregular_object_crop_box_filter_node_param_path
      • irregular_object_scan_ground_filter_node_param_path
      • irregular_object_roi_pointcloud_fusion_node_param_path
      • irregular_object_fusion_camera_param_path
      • sync_param_path
  • launch/object_recognition/detection/detector/camera_vru_detector.launch.xml
      • input/camera0/info [default: /sensing/camera/camera0/camera_info]
      • input/camera0/rois [default: /perception/object_recognition/detection/rois0]
      • input/camera1/info [default: /sensing/camera/camera1/camera_info]
      • input/camera1/rois [default: /perception/object_recognition/detection/rois1]
      • input/camera2/info [default: /sensing/camera/camera2/camera_info]
      • input/camera2/rois [default: /perception/object_recognition/detection/rois2]
      • input/camera3/info [default: /sensing/camera/camera3/camera_info]
      • input/camera3/rois [default: /perception/object_recognition/detection/rois3]
      • input/camera4/info [default: /sensing/camera/camera4/camera_info]
      • input/camera4/rois [default: /perception/object_recognition/detection/rois4]
      • input/camera5/info [default: /sensing/camera/camera5/camera_info]
      • input/camera5/rois [default: /perception/object_recognition/detection/rois5]
      • input/camera6/info [default: /sensing/camera/camera6/camera_info]
      • input/camera6/rois [default: /perception/object_recognition/detection/rois6]
      • input/camera7/info [default: /sensing/camera/camera7/camera_info]
      • input/camera7/rois [default: /perception/object_recognition/detection/rois7]
      • output/objects [default: /perception/object_recognition/detection/camera_vru/objects]
      • bbox_object_locator_param_path
      • rois_ids [default: [0, 1]]
  • launch/object_recognition/detection/detector/irregular_object_detector.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • target_container
      • base_frame [default: base_link]
      • use_intra_process [default: true]
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • input/concatenation_info [default: /sensing/lidar/concatenated/pointcloud_info]
      • output/clusters
      • crop_box_filter_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/crop_box_filter.param.yaml]
      • scan_ground_filter_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/scan_ground_filter.param.yaml]
      • roi_pointcloud_fusion_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/roi_pointcloud_fusion.param.yaml]
      • fusion_camera_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/fusion_cameras.param.yaml]
      • sync_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/fusion_common.param.yaml]
  • launch/object_recognition/detection/detector/lidar_dnn_detector.launch.xml
      • lidar_detection_model_type
      • lidar_detection_model_name
      • lidar_model_param_path
      • bevfusion_model_path [default: $(var data_path)/bevfusion]
      • centerpoint_model_path [default: $(var data_path)/lidar_centerpoint]
      • centerpoint_common_param_path [default: $(find-pkg-share autoware_lidar_centerpoint)/config/centerpoint_common.param.yaml]
      • transfusion_model_path [default: $(var data_path)/lidar_transfusion]
      • use_short_range_detection [default: false]
      • lidar_short_range_detection_model_type
      • node/pointcloud_container
      • input/pointcloud
      • output/objects
      • output/short_range_objects
  • launch/object_recognition/detection/detector/lidar_rule_detector.launch.xml
      • ns
      • node/pointcloud_container
      • input/pointcloud_map/pointcloud
      • input/obstacle_segmentation/pointcloud
      • output/cluster_objects
      • output/objects
      • voxel_grid_based_euclidean_param_path
      • use_low_height_cropbox [default: false]
  • launch/object_recognition/detection/detector/lidar_semantic_segmentation_detector.launch.xml
      • euclidean_clustering/param_path
      • input/pointcloud
      • output/objects
  • launch/object_recognition/detection/detector/tracker_based_detector.launch.xml
      • detection_by_tracker_param_path
      • input/clusters
      • input/tracked_objects
      • output/objects
  • launch/object_recognition/detection/filter/object_filter.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • objects_filter_method [default: lanelet_filter]
      • object_recognition_detection_object_lanelet_filter_param_path [default: $(var perception_config_dir)/object_recognition/detection/object_filter/object_lanelet_filter.param.yaml]
      • object_recognition_detection_object_position_filter_param_path [default: $(var perception_config_dir)/object_recognition/detection/object_filter/object_position_filter.param.yaml]
      • input/objects
      • output/objects
  • launch/object_recognition/detection/filter/object_validator.launch.xml
      • objects_validation_method
      • object_recognition_detection_obstacle_pointcloud_based_validator_param_path
      • input/obstacle_pointcloud
      • input/objects
      • output/objects
  • launch/object_recognition/detection/filter/pointcloud_downsample_filter.launch.xml
      • target_container
      • use_intra_process [default: true]
      • input/pointcloud
      • output/pointcloud
      • downsample_voxel_size [default: 0.1]
  • launch/object_recognition/detection/filter/pointcloud_map_filter.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • target_container
      • use_intra_process [default: true]
      • input/pointcloud
      • output/pointcloud
      • output/downsampled/pointcloud [default: /perception/obstacle_segmentation/pointcloud_map_filtered/downsampled/pointcloud]
      • downsample_voxel_size [default: 0.1]
      • voxel_based_compare_map_filter_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/pointcloud_filter/voxel_based_compare_map_filter.param.yaml]
  • launch/object_recognition/detection/filter/radar_filter.launch.xml
      • object_sorter_param_path
      • radar_lanelet_filtering_range_param_path
      • input/radar
      • output/objects
  • launch/object_recognition/detection/merger/camera_lidar_merger.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • object_recognition_detection_object_merger_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/detection/object_merger/data_association_matrix.param.yaml]
      • object_recognition_detection_object_merger_distance_threshold_list_path [default: $(var perception_config_dir)/object_recognition/detection/object_merger/overlapped_judge.param.yaml]
      • ml_camera_lidar_object_association_merger_param_path [default: $(var perception_config_dir)/object_recognition/detection/object_merger/ml_camera_lidar_object_association_merger.param.yaml]
      • roi_detected_object_fusion_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/roi_detected_object_fusion.param.yaml]
      • sync_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/fusion_common.param.yaml]
      • lidar_detection_model_type
      • use_detection_by_tracker
      • use_irregular_object_detector
      • use_roi_detected_object_validator
      • euclidean_clustering/shape_policy
      • use_object_filter
      • objects_filter_method
      • number_of_cameras
      • input/camera0/image
      • input/camera0/info
      • input/camera0/rois
      • input/camera1/image
      • input/camera1/info
      • input/camera1/rois
      • input/camera2/image
      • input/camera2/info
      • input/camera2/rois
      • input/camera3/image
      • input/camera3/info
      • input/camera3/rois
      • input/camera4/image
      • input/camera4/info
      • input/camera4/rois
      • input/camera5/image
      • input/camera5/info
      • input/camera5/rois
      • input/camera6/image
      • input/camera6/info
      • input/camera6/rois
      • input/camera7/image
      • input/camera7/info
      • input/camera7/rois
      • input/camera8/image
      • input/camera8/info
      • input/camera8/rois
      • input/concatenation_info [default: /sensing/lidar/concatenated/pointcloud_info]
      • input/lidar_ml/objects
      • input/lidar_cluster/objects
      • input/irregular_object/objects [default: ]
      • input/detection_by_tracker/objects
      • output/objects [default: objects]
      • alpha_merger_priority_mode [default: 0]
  • launch/object_recognition/object_recognition_camera.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • use_camera_vru_detector [default: false]
      • camera_vru_detector_rois_ids [default: [0]]
      • use_camera_bev_detector [default: true]
      • use_vector_map [default: true]
      • tracker_publish_merged_objects [default: false]
      • input/camera0/image [default: /sensing/camera/camera0/image_rect_color]
      • input/camera0/info [default: /sensing/camera/camera0/camera_info]
      • input/camera0/rois [default: /perception/object_recognition/detection/rois0]
      • input/camera1/image [default: /sensing/camera/camera1/image_rect_color]
      • input/camera1/info [default: /sensing/camera/camera1/camera_info]
      • input/camera1/rois [default: /perception/object_recognition/detection/rois1]
      • input/camera2/image [default: /sensing/camera/camera2/image_rect_color]
      • input/camera2/info [default: /sensing/camera/camera2/camera_info]
      • input/camera2/rois [default: /perception/object_recognition/detection/rois2]
      • input/camera3/image [default: /sensing/camera/camera3/image_rect_color]
      • input/camera3/info [default: /sensing/camera/camera3/camera_info]
      • input/camera3/rois [default: /perception/object_recognition/detection/rois3]
      • input/camera4/image [default: /sensing/camera/camera4/image_rect_color]
      • input/camera4/info [default: /sensing/camera/camera4/camera_info]
      • input/camera4/rois [default: /perception/object_recognition/detection/rois4]
      • input/camera5/image [default: /sensing/camera/camera5/image_rect_color]
      • input/camera5/info [default: /sensing/camera/camera5/camera_info]
      • input/camera5/rois [default: /perception/object_recognition/detection/rois5]
      • input/camera6/image [default: /sensing/camera/camera6/image_rect_color]
      • input/camera6/info [default: /sensing/camera/camera6/camera_info]
      • input/camera6/rois [default: /perception/object_recognition/detection/rois6]
      • input/camera7/image [default: /sensing/camera/camera7/image_rect_color]
      • input/camera7/info [default: /sensing/camera/camera7/camera_info]
      • input/camera7/rois [default: /perception/object_recognition/detection/rois7]
      • output/tracker_merged_objects [default: /perception/object_recognition/detection/objects]
      • output/tracked_objects [default: /perception/object_recognition/tracking/objects]
      • output/predicted_objects [default: /perception/object_recognition/objects]
      • near_range_camera_vru_param_path [default: $(var perception_config_dir)/object_recognition/detection/camera_vru_detection/near_range_camera_vru_detector.param.yaml]
      • object_recognition_tracking_multi_object_tracker_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/data_association_matrix.param.yaml]
      • object_recognition_tracking_multi_object_tracker_input_channels_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/input_channels.param.yaml]
      • object_recognition_tracking_multi_object_tracker_node_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/multi_object_tracker_node.param.yaml]
  • launch/object_recognition/object_recognition_camera_lidar_radar.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • lidar_detection_model [default: centerpoint]
      • use_short_range_detection [default: false]
      • lidar_short_range_detection_model_type [default: centerpoint_short_range]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • bevfusion_model_path [default: $(var data_path)/bevfusion]
      • pointpainting_model_path [default: $(var data_path)/image_projection_based_fusion]
      • camera_2d_detector/model_path [default: $(var data_path)/tensorrt_yolox/yolox-sPlus-opt-pseudoV2-T4-960x960-T4-seg16cls.onnx]
      • camera_2d_detector/label_path [default: $(var data_path)/tensorrt_yolox/label.txt]
      • camera_2d_detector/color_map_path [default: $(var data_path)/tensorrt_yolox/semseg_color_map.csv]
      • camera_2d_detector/roi_remap_path [default: $(var perception_config_dir)/traffic_light_recognition/tensorrt_yolox/roi_label_remap.csv]
      • camera_2d_detector/roi_to_semseg_remap_path [default: $(var perception_config_dir)/traffic_light_recognition/tensorrt_yolox/roi_to_semseg_label_remap.csv]
      • use_semantic_segmentation_ptv3 [default: false]
      • use_object_filter [default: true]
      • use_pointcloud_map [default: true]
      • use_validator [default: false]
      • objects_validation_method [default: obstacle_pointcloud]
      • use_low_height_cropbox [default: false]
      • use_low_intensity_cluster_filter [default: true]
      • use_image_segmentation_based_filter [default: false]
      • use_irregular_object_detector [default: true]
      • use_camera_vru_detector [default: false]
      • camera_vru_detector_rois_ids [default: [0]]
      • use_camera_only_detection [default: false]
      • use_vector_map [default: true]
      • tracker_publish_merged_objects [default: false]
      • number_of_cameras [default: 6]
      • pointcloud_container_name [default: pointcloud_container]
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • input/semantic_segmentation/pointcloud [default: /perception/segmented/pointcloud]
      • input/obstacle_segmentation/pointcloud [default: /perception/obstacle_segmentation/pointcloud]
      • input/camera0/image [default: /sensing/camera/camera0/image_rect_color]
      • input/camera0/info [default: /sensing/camera/camera0/camera_info]
      • input/camera0/rois [default: /perception/object_recognition/detection/rois0]
      • input/camera1/image [default: /sensing/camera/camera1/image_rect_color]
      • input/camera1/info [default: /sensing/camera/camera1/camera_info]
      • input/camera1/rois [default: /perception/object_recognition/detection/rois1]
      • input/camera2/image [default: /sensing/camera/camera2/image_rect_color]
      • input/camera2/info [default: /sensing/camera/camera2/camera_info]
      • input/camera2/rois [default: /perception/object_recognition/detection/rois2]
      • input/camera3/image [default: /sensing/camera/camera3/image_rect_color]
      • input/camera3/info [default: /sensing/camera/camera3/camera_info]
      • input/camera3/rois [default: /perception/object_recognition/detection/rois3]
      • input/camera4/image [default: /sensing/camera/camera4/image_rect_color]
      • input/camera4/info [default: /sensing/camera/camera4/camera_info]
      • input/camera4/rois [default: /perception/object_recognition/detection/rois4]
      • input/camera5/image [default: /sensing/camera/camera5/image_rect_color]
      • input/camera5/info [default: /sensing/camera/camera5/camera_info]
      • input/camera5/rois [default: /perception/object_recognition/detection/rois5]
      • input/camera6/image [default: /sensing/camera/camera6/image_rect_color]
      • input/camera6/info [default: /sensing/camera/camera6/camera_info]
      • input/camera6/rois [default: /perception/object_recognition/detection/rois6]
      • input/camera7/image [default: /sensing/camera/camera7/image_rect_color]
      • input/camera7/info [default: /sensing/camera/camera7/camera_info]
      • input/camera7/rois [default: /perception/object_recognition/detection/rois7]
      • input/camera8/image [default: /sensing/camera/camera8/image_rect_color]
      • input/camera8/info [default: /sensing/camera/camera8/camera_info]
      • input/camera8/rois [default: /perception/object_recognition/detection/rois8]
      • input/concatenation_info [default: /sensing/lidar/concatenated/pointcloud_info]
      • input/radar [default: /sensing/radar/detected_objects]
      • input/camera_only/objects [default: /perception/object_recognition/detection/camera_only/objects]
      • input/camera_only/channel [default: camera_streampetr]
      • image_topic_name [default: image_rect_color]
      • segmentation_pointcloud_fusion_camera_ids [default: [0,2,4]]
      • output/tracker_merged_objects [default: /perception/object_recognition/detection/objects]
      • output/tracked_objects [default: /perception/object_recognition/tracking/objects]
      • output/predicted_objects [default: /perception/object_recognition/objects]
      • sync_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/fusion_common.param.yaml]
      • roi_cluster_fusion_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/roi_cluster_fusion.param.yaml]
      • pointpainting_fusion_common_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/pointpainting_common.param.yaml]
      • irregular_object_crop_box_filter_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/crop_box_filter.param.yaml]
      • irregular_object_scan_ground_filter_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/scan_ground_filter.param.yaml]
      • irregular_object_roi_pointcloud_fusion_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/roi_pointcloud_fusion.param.yaml]
      • irregular_object_fusion_camera_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/fusion_cameras.param.yaml]
      • lidar_model_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model]
      • centerpoint_common_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model/centerpoint_common.param.yaml]
      • voxel_grid_based_euclidean_param_path [default: $(var perception_config_dir)/object_recognition/detection/clustering/voxel_grid_based_euclidean_cluster.param.yaml]
      • radar_lanelet_filtering_range_param_path [default: $(var perception_config_dir)/object_recognition/detection/object_filter/radar_lanelet_filter.param.yaml]
      • object_recognition_detection_object_sorter_radar_param_path [default: $(var perception_config_dir)/object_recognition/detection/radar_filter/detected_object_sorter.param.yaml]
      • near_range_camera_vru_param_path [default: $(var perception_config_dir)/object_recognition/detection/camera_vru_detection/near_range_camera_vru_detector.param.yaml]
      • object_recognition_detection_obstacle_pointcloud_based_validator_param_path [default: $(var perception_config_dir)/object_recognition/detection/detected_object_validation/obstacle_pointcloud_based_validator.param.yaml]
      • object_recognition_tracking_multi_object_tracker_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/data_association_matrix.param.yaml]
      • object_recognition_tracking_multi_object_tracker_input_channels_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/input_channels.param.yaml]
      • object_recognition_tracking_multi_object_tracker_node_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/multi_object_tracker_node.param.yaml]
      • euclidean_clustering/param_path [default: $(var perception_config_dir)/object_recognition/detection/euclidean_cluster/label_based_euclidean_cluster.param.yaml]
  • launch/object_recognition/object_recognition_camera_lidar_radar_serial.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • lidar_detection_model [default: centerpoint]
      • use_short_range_detection [default: false]
      • lidar_short_range_detection_model_type [default: centerpoint_short_range]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • bevfusion_model_path [default: $(var data_path)/bevfusion]
      • pointpainting_model_path [default: $(var data_path)/image_projection_based_fusion]
      • camera_2d_detector/model_path [default: $(var data_path)/tensorrt_yolox/yolox-sPlus-opt-pseudoV2-T4-960x960-T4-seg16cls.onnx]
      • camera_2d_detector/label_path [default: $(var data_path)/tensorrt_yolox/label.txt]
      • camera_2d_detector/color_map_path [default: $(var data_path)/tensorrt_yolox/semseg_color_map.csv]
      • camera_2d_detector/roi_remap_path [default: $(var perception_config_dir)/traffic_light_recognition/tensorrt_yolox/roi_label_remap.csv]
      • camera_2d_detector/roi_to_semseg_remap_path [default: $(var perception_config_dir)/traffic_light_recognition/tensorrt_yolox/roi_to_semseg_label_remap.csv]
      • use_object_filter [default: true]
      • objects_filter_method [default: lanelet_filter]
      • use_pointcloud_map [default: true]
      • use_validator [default: false]
      • objects_validation_method [default: obstacle_pointcloud]
      • use_roi_detected_object_validator [default: true]
      • use_detection_by_tracker [default: true]
      • use_low_height_cropbox [default: false]
      • use_low_intensity_cluster_filter [default: true]
      • use_image_segmentation_based_filter [default: false]
      • use_irregular_object_detector [default: true]
      • ml_camera_lidar_merger_priority_mode [default: 0]
      • use_vector_map [default: true]
      • number_of_cameras [default: 6]
      • pointcloud_container_name [default: pointcloud_container]
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • input/obstacle_segmentation/pointcloud [default: /perception/obstacle_segmentation/pointcloud]
      • input/camera0/image [default: /sensing/camera/camera0/image_rect_color]
      • input/camera0/info [default: /sensing/camera/camera0/camera_info]
      • input/camera0/rois [default: /perception/object_recognition/detection/rois0]
      • input/camera1/image [default: /sensing/camera/camera1/image_rect_color]
      • input/camera1/info [default: /sensing/camera/camera1/camera_info]
      • input/camera1/rois [default: /perception/object_recognition/detection/rois1]
      • input/camera2/image [default: /sensing/camera/camera2/image_rect_color]
      • input/camera2/info [default: /sensing/camera/camera2/camera_info]
      • input/camera2/rois [default: /perception/object_recognition/detection/rois2]
      • input/camera3/image [default: /sensing/camera/camera3/image_rect_color]
      • input/camera3/info [default: /sensing/camera/camera3/camera_info]
      • input/camera3/rois [default: /perception/object_recognition/detection/rois3]
      • input/camera4/image [default: /sensing/camera/camera4/image_rect_color]
      • input/camera4/info [default: /sensing/camera/camera4/camera_info]
      • input/camera4/rois [default: /perception/object_recognition/detection/rois4]
      • input/camera5/image [default: /sensing/camera/camera5/image_rect_color]
      • input/camera5/info [default: /sensing/camera/camera5/camera_info]
      • input/camera5/rois [default: /perception/object_recognition/detection/rois5]
      • input/camera6/image [default: /sensing/camera/camera6/image_rect_color]
      • input/camera6/info [default: /sensing/camera/camera6/camera_info]
      • input/camera6/rois [default: /perception/object_recognition/detection/rois6]
      • input/camera7/image [default: /sensing/camera/camera7/image_rect_color]
      • input/camera7/info [default: /sensing/camera/camera7/camera_info]
      • input/camera7/rois [default: /perception/object_recognition/detection/rois7]
      • input/camera8/image [default: /sensing/camera/camera8/image_rect_color]
      • input/camera8/info [default: /sensing/camera/camera8/camera_info]
      • input/camera8/rois [default: /perception/object_recognition/detection/rois8]
      • input/concatenation_info [default: /sensing/lidar/concatenated/pointcloud_info]
      • input/radar/tracked_objects [default: /sensing/radar/tracked_objects]
      • image_topic_name [default: image_rect_color]
      • segmentation_pointcloud_fusion_camera_ids [default: [0,2,4]]
      • output/tracked_objects [default: /perception/object_recognition/tracking/objects]
      • output/predicted_objects [default: /perception/object_recognition/objects]
      • sync_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/fusion_common.param.yaml]
      • roi_cluster_fusion_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/roi_cluster_fusion.param.yaml]
      • pointpainting_fusion_common_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/pointpainting_common.param.yaml]
      • irregular_object_crop_box_filter_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/crop_box_filter.param.yaml]
      • irregular_object_scan_ground_filter_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/scan_ground_filter.param.yaml]
      • irregular_object_roi_pointcloud_fusion_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/roi_pointcloud_fusion.param.yaml]
      • irregular_object_fusion_camera_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/fusion_cameras.param.yaml]
      • lidar_model_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model]
      • centerpoint_common_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model/centerpoint_common.param.yaml]
      • voxel_grid_based_euclidean_param_path [default: $(var perception_config_dir)/object_recognition/detection/clustering/voxel_grid_based_euclidean_cluster.param.yaml]
      • detection_by_tracker_param_path [default: $(var perception_config_dir)/object_recognition/detection/detection_by_tracker/detection_by_tracker.param.yaml]
      • object_recognition_detection_obstacle_pointcloud_based_validator_param_path [default: $(var perception_config_dir)/object_recognition/detection/detected_object_validation/obstacle_pointcloud_based_validator.param.yaml]
      • object_recognition_detection_object_merger_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/detection/object_merger/data_association_matrix.param.yaml]
      • object_recognition_detection_object_merger_distance_threshold_list_path [default: $(var perception_config_dir)/object_recognition/detection/object_merger/overlapped_judge.param.yaml]
      • ml_camera_lidar_object_association_merger_param_path [default: $(var perception_config_dir)/object_recognition/detection/object_merger/ml_camera_lidar_object_association_merger.param.yaml]
      • roi_detected_object_fusion_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/roi_detected_object_fusion.param.yaml]
      • object_recognition_tracking_multi_object_tracker_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/data_association_matrix.param.yaml]
      • object_recognition_tracking_multi_object_tracker_input_channels_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/input_channels.param.yaml]
      • object_recognition_tracking_multi_object_tracker_node_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/multi_object_tracker_node.param.yaml]
      • object_recognition_tracking_radar_tracked_object_sorter_param_path [default: $(var perception_config_dir)/object_recognition/tracking/radar_object_tracker/tracked_object_sorter.param.yaml]
      • object_recognition_tracking_radar_tracked_object_lanelet_filter_param_path [default: $(var perception_config_dir)/object_recognition/tracking/radar_object_tracker/tracked_object_lanelet_filter.param.yaml]
      • object_recognition_tracking_object_merger_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/tracking/tracking_object_merger/data_association_matrix.param.yaml]
      • object_recognition_tracking_object_merger_node_param_path [default: $(var perception_config_dir)/object_recognition/tracking/tracking_object_merger/decorative_tracker_merger.param.yaml]
      • euclidean_clustering/shape_policy [default: 0]
  • launch/object_recognition/object_recognition_lidar.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • lidar_detection_model [default: centerpoint]
      • use_short_range_detection [default: false]
      • lidar_short_range_detection_model_type [default: centerpoint_short_range]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • bevfusion_model_path [default: $(var data_path)/bevfusion]
      • use_pointcloud_map [default: true]
      • use_validator [default: false]
      • objects_validation_method [default: obstacle_pointcloud]
      • use_object_filter [default: true]
      • objects_filter_method [default: lanelet_filter]
      • use_low_height_cropbox [default: false]
      • use_vector_map [default: true]
      • tracker_publish_merged_objects [default: false]
      • pointcloud_container_name [default: pointcloud_container]
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • input/obstacle_segmentation/pointcloud [default: /perception/obstacle_segmentation/pointcloud]
      • output/tracker_merged_objects [default: /perception/object_recognition/detection/objects]
      • output/tracked_objects [default: /perception/object_recognition/tracking/objects]
      • output/predicted_objects [default: /perception/object_recognition/objects]
      • lidar_model_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model]
      • centerpoint_common_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model/centerpoint_common.param.yaml]
      • voxel_grid_based_euclidean_param_path [default: $(var perception_config_dir)/object_recognition/detection/clustering/voxel_grid_based_euclidean_cluster.param.yaml]
      • object_recognition_detection_obstacle_pointcloud_based_validator_param_path [default: $(var perception_config_dir)/object_recognition/detection/detected_object_validation/obstacle_pointcloud_based_validator.param.yaml]
      • object_recognition_tracking_multi_object_tracker_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/data_association_matrix.param.yaml]
      • object_recognition_tracking_multi_object_tracker_input_channels_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/input_channels.param.yaml]
      • object_recognition_tracking_multi_object_tracker_node_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/multi_object_tracker_node.param.yaml]
  • launch/object_recognition/object_recognition_lidar_ptv3.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • lidar_detection_model [default: centerpoint]
      • use_short_range_detection [default: false]
      • lidar_short_range_detection_model_type [default: centerpoint_short_range]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • bevfusion_model_path [default: $(var data_path)/bevfusion]
      • use_vector_map [default: true]
      • tracker_publish_merged_objects [default: false]
      • pointcloud_container_name [default: pointcloud_container]
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • input/semantic_segmentation/pointcloud [default: /perception/segmented/pointcloud]
      • output/tracker_merged_objects [default: /perception/object_recognition/detection/objects]
      • output/tracked_objects [default: /perception/object_recognition/tracking/objects]
      • output/predicted_objects [default: /perception/object_recognition/objects]
      • lidar_model_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model]
      • centerpoint_common_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model/centerpoint_common.param.yaml]
      • object_recognition_tracking_multi_object_tracker_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/data_association_matrix.param.yaml]
      • object_recognition_tracking_multi_object_tracker_input_channels_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/input_channels.param.yaml]
      • object_recognition_tracking_multi_object_tracker_node_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/multi_object_tracker_node.param.yaml]
      • euclidean_clustering/param_path [default: $(var perception_config_dir)/object_recognition/detection/euclidean_cluster/label_based_euclidean_cluster.param.yaml]
  • launch/object_recognition/object_recognition_lidar_radar.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • lidar_detection_model [default: centerpoint]
      • use_short_range_detection [default: false]
      • lidar_short_range_detection_model_type [default: centerpoint_short_range]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • bevfusion_model_path [default: $(var data_path)/bevfusion]
      • use_low_height_cropbox [default: false]
      • use_vector_map [default: true]
      • tracker_publish_merged_objects [default: false]
      • pointcloud_container_name [default: pointcloud_container]
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • input/semantic_segmentation/pointcloud [default: /perception/segmented/pointcloud]
      • input/radar [default: /sensing/radar/detected_objects]
      • output/tracker_merged_objects [default: /perception/object_recognition/detection/objects]
      • output/tracked_objects [default: /perception/object_recognition/tracking/objects]
      • output/predicted_objects [default: /perception/object_recognition/objects]
      • lidar_model_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model]
      • centerpoint_common_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model/centerpoint_common.param.yaml]
      • voxel_grid_based_euclidean_param_path [default: $(var perception_config_dir)/object_recognition/detection/clustering/voxel_grid_based_euclidean_cluster.param.yaml]
      • radar_lanelet_filtering_range_param_path [default: $(var perception_config_dir)/object_recognition/detection/object_filter/radar_lanelet_filter.param.yaml]
      • object_recognition_detection_object_sorter_radar_param_path [default: $(var perception_config_dir)/object_recognition/detection/radar_filter/detected_object_sorter.param.yaml]
      • object_recognition_tracking_multi_object_tracker_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/data_association_matrix.param.yaml]
      • object_recognition_tracking_multi_object_tracker_input_channels_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/input_channels.param.yaml]
      • object_recognition_tracking_multi_object_tracker_node_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/multi_object_tracker_node.param.yaml]
      • euclidean_clustering/param_path [default: $(var perception_config_dir)/object_recognition/detection/euclidean_cluster/label_based_euclidean_cluster.param.yaml]
  • launch/object_recognition/prediction/prediction.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • use_vector_map [default: false]
      • prediction_model_type [default: map_based]
      • input/objects [default: /perception/object_recognition/tracking/objects]
      • output/objects [default: /perception/object_recognition/objects]
      • object_recognition_prediction_map_based_prediction_param_path [default: $(var perception_config_dir)/object_recognition/prediction/map_based_prediction.param.yaml]
  • launch/obstacle_segmentation/ground_segmentation_additional_lidars.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • target_container
      • lidar_name
      • base_frame [default: base_link]
      • use_intra_process [default: true]
      • input/main/pointcloud
      • input/lidar/pointcloud [default: /sensing/lidar/$(var lidar_name)/pointcloud]
      • output/pointcloud
      • output/pointcloud_info [default: $(var output/pointcloud)_info]
      • crop_box_margin_max_z [default: 2.5]
      • crop_box_margin_min_z [default: -2.5]
      • crop_box_filter_node_param_path [default: $(var perception_config_dir)/obstacle_segmentation/ground_segmentation/crop_box_filter.param.yaml]
      • scan_ground_filter_node_param_path [default: $(var perception_config_dir)/obstacle_segmentation/ground_segmentation/scan_ground_filter.param.yaml]
  • launch/obstacle_segmentation/ground_segmentation_ml.launch.xml
      • pointcloud_container_name [default: pointcloud_container]
      • use_pointcloud_container [default: true]
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • output/pointcloud [default: /perception/obstacle_segmentation/pointcloud]
      • output/pointcloud/segmentation [default: /perception/segmented/pointcloud]
      • output/pointcloud/visualization [default: /perception/segmented/pointcloud/visualization]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • model_variant [default: ptv3]
      • model_path [default: $(var data_path)/ptv3]
  • launch/obstacle_segmentation/ground_segmentation_ransac.launch.xml
      • target_container
      • base_frame [default: base_link]
      • use_intra_process [default: true]
      • input_topics
      • input/main/pointcloud
      • output/pointcloud
      • output/pointcloud_info [default: $(var output/pointcloud)_info]
      • detection_area_crop_box_filter_node_param_path
      • ransac_ground_filter_node_param_path [default: $(find-pkg-share autoware_ground_segmentation)/config/ransac_ground_filter.param.yaml]
  • launch/obstacle_segmentation/ground_segmentation_rule_based.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • target_container
      • base_frame [default: base_link]
      • use_intra_process [default: true]
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • output/pointcloud [default: /perception/obstacle_segmentation/pointcloud]
      • crop_box_margin_max_z [default: 2.5]
      • crop_box_margin_min_z [default: -2.5]
      • crop_box_filter_node_param_path [default: $(var perception_config_dir)/obstacle_segmentation/ground_segmentation/crop_box_filter.param.yaml]
      • scan_ground_filter_node_param_path [default: $(var perception_config_dir)/obstacle_segmentation/ground_segmentation/scan_ground_filter.param.yaml]
  • launch/obstacle_segmentation/ground_segmentation_rule_based_cuda.launch.xml
      • target_container
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • output/pointcloud [default: /perception/obstacle_segmentation/pointcloud]
      • output/ground_pointcloud [default: /perception/obstacle_segmentation/ground_pointcloud]
      • cuda_scan_ground_segmentation_filter_node_param_path [default: $(find-pkg-share autoware_ground_segmentation_cuda)/config/cuda_scan_ground_segmentation_filter.param.yaml]
  • launch/occupancy_grid_map/probabilistic_occupancy_grid_map.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • input/obstacle_pointcloud [default: concatenated/pointcloud]
      • input/raw_pointcloud [default: no_ground/oneshot/pointcloud]
      • output [default: /perception/occupancy_grid_map/map]
      • use_intra_process [default: false]
      • use_multithread [default: false]
      • pointcloud_container_name [default: pointcloud_container]
      • occupancy_grid_map_method
      • occupancy_grid_map_param_path [default: $(var perception_config_dir)/occupancy_grid_map/$(var occupancy_grid_map_method).param.yaml]
      • occupancy_grid_map_updater
      • occupancy_grid_map_updater_param_path [default: $(var perception_config_dir)/occupancy_grid_map/$(var occupancy_grid_map_updater)_updater.param.yaml]
      • input_obstacle_pointcloud [default: false]
      • input_obstacle_and_raw_pointcloud [default: true]
      • use_pointcloud_container [default: true]
  • launch/traffic_light_recognition/traffic_light_detector_fine.launch.xml
    • Fine-detection chain for one traffic light camera: map based detector proposes rough ROIs, the fine detector refines them, classifiers label them.
      • perception_config_pkg [default: autoware_perception_config]
      • camera_namespace
      • enable_image_decompressor [default: true]
      • input/vector_map [default: /map/vector_map]
      • input/route [default: /planning/mission_planning/route]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • map_based_detector_param_path [default: $(find-pkg-share autoware_traffic_light_map_based_detector)/config/traffic_light_map_based_detector.param.yaml]
      • fine_detector_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_fine_detector/traffic_light_fine_detector.param.yaml]
      • car_classifier_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_classifier/car_traffic_light_classifier.param.yaml]
      • pedestrian_classifier_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_classifier/pedestrian_traffic_light_classifier.param.yaml]
      • lamp_recognizer_ml_param_path [default: $(var data_path)/traffic_light_classifier/lamp_recognizer_ml.param.yaml]
      • roi_visualizer_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_visualization/traffic_light_roi_visualizer.param.yaml]
      • fine_detector_model_path [default: $(var data_path)/traffic_light_fine_detector/tlr_car_ped_yolox_s_batch_6.onnx]
      • fine_detector_label_path [default: $(var data_path)/traffic_light_fine_detector/tlr_labels.txt]
      • car_classifier_model_path [default: $(var data_path)/traffic_light_classifier/traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • car_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels.txt]
      • pedestrian_classifier_model_path [default: $(var data_path)/traffic_light_classifier/ped_traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • pedestrian_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels_ped.txt]
      • car_classifier_type [default: 1]
      • pedestrian_classifier_type [default: 1]
  • launch/traffic_light_recognition/traffic_light_detector_rough.launch.xml
    • Rough-detection chain for one traffic light camera: map based detector ROIs feed the classifiers directly, without a high accuracy detector.
      • perception_config_pkg [default: autoware_perception_config]
      • camera_namespace
      • enable_image_decompressor [default: true]
      • input/vector_map [default: /map/vector_map]
      • input/route [default: /planning/mission_planning/route]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • map_based_detector_param_path [default: $(find-pkg-share autoware_traffic_light_map_based_detector)/config/traffic_light_map_based_detector.param.yaml]
      • car_classifier_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_classifier/car_traffic_light_classifier.param.yaml]
      • pedestrian_classifier_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_classifier/pedestrian_traffic_light_classifier.param.yaml]
      • lamp_recognizer_ml_param_path [default: $(var data_path)/traffic_light_classifier/lamp_recognizer_ml.param.yaml]
      • roi_visualizer_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_visualization/traffic_light_roi_visualizer.param.yaml]
      • car_classifier_model_path [default: $(var data_path)/traffic_light_classifier/traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • car_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels.txt]
      • pedestrian_classifier_model_path [default: $(var data_path)/traffic_light_classifier/ped_traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • pedestrian_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels_ped.txt]
      • car_classifier_type [default: 1]
      • pedestrian_classifier_type [default: 1]
  • launch/traffic_light_recognition/traffic_light_detector_whole_image.launch.xml
    • Whole-image detection chain for one traffic light camera: yolox detects over the full image, the selector matches detections to map based ROIs, classifiers label them.
      • perception_config_pkg [default: autoware_perception_config]
      • camera_namespace
      • enable_image_decompressor [default: true]
      • input/vector_map [default: /map/vector_map]
      • input/route [default: /planning/mission_planning/route]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • map_based_detector_param_path [default: $(find-pkg-share autoware_traffic_light_map_based_detector)/config/traffic_light_map_based_detector.param.yaml]
      • whole_image_detector_param_path [default: $(var perception_config_dir)/traffic_light_recognition/tensorrt_yolox/yolox_traffic_light_detector.param.yaml]
      • car_classifier_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_classifier/car_traffic_light_classifier.param.yaml]
      • pedestrian_classifier_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_classifier/pedestrian_traffic_light_classifier.param.yaml]
      • lamp_recognizer_ml_param_path [default: $(var data_path)/traffic_light_classifier/lamp_recognizer_ml.param.yaml]
      • roi_visualizer_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_visualization/traffic_light_roi_visualizer.param.yaml]
      • whole_image_detector_model_path [default: $(var data_path)/tensorrt_yolox/yolox_s_car_ped_tl_detector_960_960_batch_1.onnx]
      • whole_image_detector_label_path [default: $(var data_path)/tensorrt_yolox/car_ped_tl_detector_labels.txt]
      • car_classifier_model_path [default: $(var data_path)/traffic_light_classifier/traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • car_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels.txt]
      • pedestrian_classifier_model_path [default: $(var data_path)/traffic_light_classifier/ped_traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • pedestrian_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels_ped.txt]
      • car_classifier_type [default: 1]
      • pedestrian_classifier_type [default: 1]
  • launch/traffic_light_recognition/traffic_light_fusion.launch.xml
    • Post-process shared by all traffic light cameras: multi camera fusion, V2X arbitration, pedestrian signal estimation and map visualization.
      • perception_config_pkg [default: autoware_perception_config]
      • camera_namespaces [default: [camera6, camera7]]
      • input/vector_map [default: /map/vector_map]
      • external/traffic_signals [default: /perception/traffic_light_recognition/external/traffic_signals]
      • internal/traffic_signals [default: /perception/traffic_light_recognition/internal/traffic_signals]
      • judged/traffic_signals [default: /perception/traffic_light_recognition/judged/traffic_signals]
      • output/traffic_signals [default: /perception/traffic_light_recognition/traffic_signals]
      • multi_camera_fusion_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_multi_camera_fusion/traffic_light_multi_camera_fusion.param.yaml]
      • arbiter_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_arbiter/traffic_light_arbiter.param.yaml]
      • crosswalk_estimator_param_path [default: $(var perception_config_dir)/traffic_light_recognition/crosswalk_traffic_light_estimator/crosswalk_traffic_light_estimator.param.yaml]
  • launch/traffic_light_recognition/traffic_light_occlusion_predictor.launch.xml
    • Occlusion check for one traffic light camera: merges the car and pedestrian signals into the per-camera output, masking signals hidden from the camera.
      • perception_config_pkg [default: autoware_perception_config]
      • camera_namespace
      • input/cloud [default: /sensing/lidar/top/pointcloud_raw_ex]
      • input/vector_map [default: /map/vector_map]
      • occlusion_predictor_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_occlusion_predictor/traffic_light_occlusion_predictor.param.yaml]
  • launch/traffic_light_recognition/traffic_light_recognition_fine.launch.xml
    • Traffic light recognition, fine detection mode: per-camera fine detection with occlusion check, then shared fusion. Two cameras as the representative set; add or remove per-camera blocks to match the vehicle.
      • perception_config_pkg [default: autoware_perception_config]
      • input/vector_map [default: /map/vector_map]
      • input/route [default: /planning/mission_planning/route]
      • input/pointcloud [default: /sensing/lidar/top/pointcloud_raw_ex]
      • external/traffic_signals [default: /perception/traffic_light_recognition/external/traffic_signals]
      • output/traffic_signals [default: /perception/traffic_light_recognition/traffic_signals]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • fine_detector_model_path [default: $(var data_path)/traffic_light_fine_detector/tlr_car_ped_yolox_s_batch_6.onnx]
      • fine_detector_label_path [default: $(var data_path)/traffic_light_fine_detector/tlr_labels.txt]
      • lamp_recognizer_ml_param_path [default: $(var data_path)/traffic_light_classifier/lamp_recognizer_ml.param.yaml]
      • car_classifier_model_path [default: $(var data_path)/traffic_light_classifier/traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • car_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels.txt]
      • pedestrian_classifier_model_path [default: $(var data_path)/traffic_light_classifier/ped_traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • pedestrian_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels_ped.txt]
      • car_classifier_type [default: 1]
      • pedestrian_classifier_type [default: 1]
      • base_map_based_detector_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_map_based_detector/base_traffic_light_map_based_detector.param.yaml]
      • extra_map_based_detector_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_map_based_detector/extra_traffic_light_map_based_detector.param.yaml]
  • launch/traffic_light_recognition/traffic_light_recognition_merge.launch.xml
    • Traffic light recognition, merge mode: detection and classification run on other ECUs; this side runs the per-camera occlusion check and the shared fusion. Two cameras as the representative set; add or remove per-camera blocks to match the vehicle.
      • input/vector_map [default: /map/vector_map]
      • input/pointcloud [default: /sensing/lidar/top/pointcloud_raw_ex]
      • external/traffic_signals [default: /perception/traffic_light_recognition/external/traffic_signals]
      • output/traffic_signals [default: /perception/traffic_light_recognition/traffic_signals]
  • launch/traffic_light_recognition/traffic_light_recognition_rough.launch.xml
    • Traffic light recognition, rough detection mode: per-camera map based detection with occlusion check, then shared fusion. One camera as the representative set; add or remove per-camera blocks to match the vehicle.
      • perception_config_pkg [default: autoware_perception_config]
      • input/vector_map [default: /map/vector_map]
      • input/route [default: /planning/mission_planning/route]
      • input/pointcloud [default: /sensing/lidar/top/pointcloud_raw_ex]
      • external/traffic_signals [default: /perception/traffic_light_recognition/external/traffic_signals]
      • output/traffic_signals [default: /perception/traffic_light_recognition/traffic_signals]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • lamp_recognizer_ml_param_path [default: $(var data_path)/traffic_light_classifier/lamp_recognizer_ml.param.yaml]
      • car_classifier_model_path [default: $(var data_path)/traffic_light_classifier/traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • car_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels.txt]
      • pedestrian_classifier_model_path [default: $(var data_path)/traffic_light_classifier/ped_traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • pedestrian_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels_ped.txt]
      • car_classifier_type [default: 1]
      • pedestrian_classifier_type [default: 1]
      • base_map_based_detector_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_map_based_detector/base_traffic_light_map_based_detector.param.yaml]
  • launch/traffic_light_recognition/traffic_light_recognition_whole_image.launch.xml
    • Traffic light recognition, whole-image detection mode: per-camera yolox detection, then shared fusion. Two cameras as the representative set; add or remove per-camera blocks to match the vehicle.
      • perception_config_pkg [default: autoware_perception_config]
      • input/vector_map [default: /map/vector_map]
      • input/route [default: /planning/mission_planning/route]
      • external/traffic_signals [default: /perception/traffic_light_recognition/external/traffic_signals]
      • output/traffic_signals [default: /perception/traffic_light_recognition/traffic_signals]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • whole_image_detector_model_path [default: $(var data_path)/tensorrt_yolox/yolox_s_car_ped_tl_detector_960_960_batch_1.onnx]
      • whole_image_detector_label_path [default: $(var data_path)/tensorrt_yolox/car_ped_tl_detector_labels.txt]
      • lamp_recognizer_ml_param_path [default: $(var data_path)/traffic_light_classifier/lamp_recognizer_ml.param.yaml]
      • car_classifier_model_path [default: $(var data_path)/traffic_light_classifier/traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • car_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels.txt]
      • pedestrian_classifier_model_path [default: $(var data_path)/traffic_light_classifier/ped_traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • pedestrian_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels_ped.txt]
      • car_classifier_type [default: 1]
      • pedestrian_classifier_type [default: 1]
      • base_map_based_detector_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_map_based_detector/base_traffic_light_map_based_detector.param.yaml]
      • extra_map_based_detector_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_map_based_detector/extra_traffic_light_map_based_detector.param.yaml]

Messages

No message files found.

Services

No service files found

Plugins

No plugins found.

Recent questions tagged autoware_perception_launch at Robotics Stack Exchange

No version for distro noetic showing github. Known supported distros are highlighted in the buttons above.
Package symbol

autoware_perception_launch package from autoware_launch repo

autoware_launch autoware_sample_designs autoware_control_config autoware_control_launch autoware_localization_config autoware_localization_launch autoware_map_config autoware_map_launch autoware_perception_config autoware_perception_launch autoware_planning_config autoware_planning_launch autoware_sensing_launch autoware_sensor_driver_launch awsim_labs_sensor_kit_description awsim_labs_sensor_kit_launch awsim_sensor_kit_description awsim_sensor_kit_launch carla_sensor_kit_description carla_sensor_kit_launch sample_sensor_kit_description sample_sensor_kit_launch single_lidar_sensor_kit_description single_lidar_sensor_kit_launch tier4_autoware_api_launch tier4_simulator_launch tier4_system_launch tier4_vehicle_launch awsim_labs_vehicle_description awsim_labs_vehicle_launch sample_vehicle_description sample_vehicle_launch

ROS Distro
github

API Docs
Browse Code
Wiki

Package Summary

Version 0.53.0
License Apache License 2.0
Build type AMENT_CMAKE
Use RECOMMENDED

Repository Summary

Description
Checkout URI https://github.com/autowarefoundation/autoware_launch.git
VCS Type git
VCS Version main
Last Updated 2026-10-26
Dev Status UNKNOWN
Released UNRELEASED
Contributing Help Wanted (-)
Good First Issues (-)
Pull Requests to Review (-)

Package Description

The autoware_perception_launch package

Maintainers

  • Yukihiro Saito
  • Dai Nguyen
  • Taekjin Lee
  • Yoshi Ri
  • Masato Saeki

Authors

No additional authors.
autoware_universe_launch/autoware_perception_launch/README.md

autoware_perception_launch

Shared perception launch library. The system component launchers live in the product layer (autoware_launch/launch/components/component_perception.launch.xml for the lidar pipeline and component_traffic_light.launch.xml for traffic light recognition); this package hosts the perception feature launchers they combine:

launch/
├── common/                        # shared preprocess feature launchers (pointcloud downsample)
├── object_recognition/
│   ├── object_recognition_<mode>.launch.xml  # detection + tracking + prediction per sensor mode
│   ├── detection/
│   │   ├── detector/              # detectors
│   │   ├── filter/                # filters
│   │   └── merger/                # detection-side serial merger (used by the camera_lidar_radar_serial launcher)
│   └── prediction/
├── obstacle_segmentation/
├── occupancy_grid_map/
└── traffic_light_recognition/
design/module/                     # Autoware System Designer modules; same subtree as launch/
├── object_recognition/
├── obstacle_segmentation/
├── occupancy_grid_map/
└── traffic_light_recognition/

Launch files

  • object_recognition/object_recognition_<mode>.launch.xml — one feature launcher per sensor mode (camera, lidar, lidar_ptv3, lidar_radar, camera_lidar_radar, camera_lidar_radar_serial); hosts the detection stage inline together with the tracker and prediction, and pushes their module namespaces. Camera and radar input topics default to their absolute sensing paths (/sensing/camera/cameraN/..., /sensing/radar/detected_objects), so callers override only what differs. Most modes wire the multi-channel tracker to the detector outputs; camera_lidar_radar_serial instead merges the detectors in the detection stage (optionally regenerating detections through detection-by-tracker), tracks the single merged detected_objects channel, and joins radar tracked objects at the decorative tracker merger. camera runs camera-only detectors (BEV and camera VRU).
  • The clustering backend is file selection, not a flag: lidar runs rule-based euclidean clustering off the obstacle segmentation cloud, while lidar_ptv3 and lidar_radar cluster the PTv3 semantic-segmentation cloud and drop the pointcloud map filter and the object validator that only the rule-based pipeline uses. Both feed their cluster objects to the tracker on the lidar_clustering channel.
  • object_recognition/detection/{detector,filter}/ — single-purpose launchers wired together by the object-recognition launchers. merger/camera_lidar_merger.launch.xml chains the detection-side object mergers for the camera_lidar_radar_serial launcher.
  • obstacle_segmentation/ground_segmentation_<backend>.launch.xml — ground removal backends (rule_based, rule_based_cuda, ml); backend selection is file selection. ground_segmentation_{additional_lidars,ransac}.launch.xml fold extra lidars into the obstacle stream for kits that carry them.
  • occupancy_grid_map/, common/, traffic_light_recognition/ — feature launchers included directly by the system components.

Namespace and container addressing

Callers own the namespaces: the system component pushes perception and object_recognition; object_recognition_<mode>.launch.xml pushes detection/tracking/prediction. The object-recognition launchers address their internal topics through the fixed /perception/object_recognition namespace. Composable nodes load into the shared pointcloud container, whose name is exposed as pointcloud_container_name and aliased to the node/pointcloud_container argument of the detector launchers.

Config

Parameter files are resolved from a config/ tree addressed through the perception_config_pkg argument, which defaults to autoware_perception_config. A product can relocate the whole tree by setting perception_config_pkg once at its entry point — the launch configuration propagates through every include — but it must then ship the same config/ layout. Products that differ in only a few files instead override the corresponding *_param_path arguments, which every feature launcher declares with a default:

<include file="$(find-pkg-share autoware_perception_launch)/launch/object_recognition/object_recognition_lidar.launch.xml">
  <arg name="perception_config_pkg" value="my_perception_config"/>
  <arg name="FOO_NODE_param_path" value="..."/>
</include>

component_perception.launch.xml, component_traffic_light.launch.xml and planning_simulator/perception_simulation.launch.xml each declare perception_config_pkg, so a product switches the tree at the component it launches.

The launch/ tree is the interface downstream distributions adopt as-is; only package.xml, CMakeLists.txt, README.md, and changelogs stay package-local. Keep changes to this tree folder-merge compatible.

Package Dependencies

Please see <exec_depend> in package.xml.

CHANGELOG

Changelog for package autoware_perception_launch

0.53.0 (2026-10-13)

  • Merge remote-tracking branch 'origin/main' into tmp/bot/bump_version_base

  • feat(autoware_launch): run crop_box_filter standalone when agnocast is enabled (#1990)

    * feat(autoware_launch): run crop_box_filter standalone when agnocast is enabled autoware_universe#13384 moves CropBoxFilterComponent to agnocast_wrapper::Node on an agnocast-only executor, which the component containers do not set up. Start each of the eight crop_box_filter nodes as a standalone process when ENABLE_AGNOCAST=1 and keep loading them into their container otherwise. The node names, namespaces, remaps and parameters are the same in both branches.

    * refactor(autoware_launch): launch crop_box_filter through the autoware_node action ---------

  • feat(autoware_perception_launch): run scan_ground_filter standalone when agnocast is enabled (#1991)

    * feat(autoware_perception_launch): run scan_ground_filter standalone when agnocast is enabled autoware_universe#13385 moves ScanGroundFilterComponent to agnocast_wrapper::Node on an agnocast-only executor, which the component containers do not set up. Start each of the three scan_ground_filter nodes as a standalone process when ENABLE_AGNOCAST=1 and keep loading them into their container otherwise. The node names, namespaces, remaps and parameters are the same in both branches.

    * refactor(autoware_perception_launch): launch scan_ground_filter through the autoware_node action ---------

  • feat(autoware_launch): run voxel_grid_downsample_filter standalone when agnocast is enabled (#1992)

    • feat(autoware_launch): run voxel_grid_downsample_filter standalone when agnocast is enabled
    • style(pre-commit): autofix

    * refactor(autoware_launch): launch voxel_grid_downsample_filter through the autoware_node action ---------Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>

  • feat(autoware_sample_designs): compose the AD API, system and vehicle components from design modules (#1976)

    * fix(control, planning): connect the operation mode state and velocity limit inputs Wires collision_detector's operation_mode_state and remaining_distance_time_calculator's planning_velocity through the module designs; their node designs now declare the fixed topic names as remap targets.

    * fix(planning): connect the internal velocity limit path to the selector Forwards motion_velocity_planner's velocity_limit and clear_velocity_limit up to ScenarioPlanning and feeds them into external_velocity_limit_selector, whose internal inputs were unconnected.

    * fix(control, planning): connect the remaining module inputs to their sources Wires the operation mode transition manager's gate and selector inputs, the stop mode operator's vehicle status, the manual lane change handler's route availability and preferred primitive service, the parking costmap, and the route marker. Exposes the vehicle control mode service, the preferred lane service and the system hazard lights command as component interfaces.

    • fix(planning): follow the hazard lights selector input rename
    • fix(planning): add steering_status subscriber to various planning modules
    • feat(vehicle): add vehicle wrapper and launch file for vehicle interface
    • style(pre-commit): autofix
    • fix(planning): add planning factors publishers and connections for various modules
    • refactor(autoware_sample_designs): compose the AD API from design modules
    • Refactor ADAPI module structure and enhance planning factors
    • Updated AdapiRvizHelpers.module.yaml to include helper prefixes for instances and connections.
    • Modified AutowareSample.system.yaml to reorganize ADAPI components, introducing default_adapi for better structure.
    • Expanded planning factors in Planning.module.yaml, LaneDriving.module.yaml, ScenarioPlanning.module.yaml, and associated modules to include additional factors for improved behavior and motion planning.
    • Enhanced connections in behavior planning and motion planning modules to accommodate new planning factors.
    • Ensured consistency across scenario planning modules by aligning

File truncated at 100 lines see the full file

Package Dependencies

Deps Name
ament_cmake_auto
autoware_cmake
autoware_agnocast_wrapper
autoware_bevfusion
autoware_bytetrack
autoware_compare_map_segmentation
autoware_crosswalk_traffic_light_estimator
autoware_cuda_pointcloud_preprocessor
autoware_detected_object_feature_remover
autoware_detected_object_validation
autoware_detection_by_tracker
autoware_elevation_map_loader
autoware_euclidean_cluster
autoware_ground_segmentation
autoware_ground_segmentation_cuda
autoware_image_object_locator
autoware_image_projection_based_fusion
autoware_image_transport_decompressor
autoware_lidar_apollo_instance_segmentation
autoware_lidar_centerpoint
autoware_lidar_transfusion
autoware_map_based_prediction
autoware_multi_object_tracker
autoware_object_merger
autoware_object_sorter
autoware_occupancy_grid_map_outlier_filter
autoware_perception_config
autoware_pointcloud_preprocessor
autoware_probabilistic_occupancy_grid_map
autoware_ptv3
autoware_raindrop_cluster_filter
autoware_shape_estimation
autoware_simpl_prediction
autoware_simple_object_merger
autoware_tensorrt_bevdet
autoware_tensorrt_yolox
autoware_tracking_object_merger
autoware_traffic_light_arbiter
autoware_traffic_light_category_merger
autoware_traffic_light_classifier
autoware_traffic_light_fine_detector
autoware_traffic_light_map_based_detector
autoware_traffic_light_multi_camera_fusion
autoware_traffic_light_occlusion_predictor
autoware_traffic_light_selector
autoware_traffic_light_visualization
rclcpp_components
topic_tools
ament_lint_auto
autoware_lint_common

System Dependencies

No direct system dependencies.

Dependant Packages

Name Deps
autoware_launch
reaction_analyzer

Launch files

  • launch/common/pointcloud_downsample.launch.xml
      • pointcloud_container_name
      • cuda_pointcloud_preprocessing [default: false]
      • input/pointcloud
      • output/pointcloud
      • common_downsample_voxel_size_x [default: 0.05]
      • common_downsample_voxel_size_y [default: 0.05]
      • common_downsample_voxel_size_z [default: 0.05]
  • launch/object_recognition/detection/detector/camera_2d_detector.launch.xml
      • image_raw0 [default: /sensing/camera/camera0/image_raw]
      • image_raw1 [default: /sensing/camera/camera1/image_raw]
      • image_raw2 [default: /sensing/camera/camera2/image_raw]
      • image_raw3 [default: /sensing/camera/camera3/image_raw]
      • image_raw4 [default: /sensing/camera/camera4/image_raw]
      • image_raw5 [default: /sensing/camera/camera5/image_raw]
      • image_raw6 [default: /sensing/camera/camera6/image_raw]
      • image_raw7 [default: /sensing/camera/camera7/image_raw]
      • image_raw8 [default: /sensing/camera/camera8/image_raw]
      • image_raw9 [default: /sensing/camera/camera9/image_raw]
      • image_number [default: 1]
      • camera_index [default: 0]
      • use_bytetrack [default: true]
      • enable_visualizer [default: false]
      • camera_2d_detector/model_path
      • camera_2d_detector/label_path
      • camera_2d_detector/color_map_path
      • camera_2d_detector/roi_remap_path
      • camera_2d_detector/roi_to_semseg_remap_path
      • tensorrt_yolox_ns [default: ]
  • launch/object_recognition/detection/detector/camera_bev_detector.launch.xml
      • input/camera0/image
      • input/camera0/info
      • input/camera1/image
      • input/camera1/info
      • input/camera2/image
      • input/camera2/info
      • input/camera3/image
      • input/camera3/info
      • input/camera4/image
      • input/camera4/info
      • input/camera5/image
      • input/camera5/info
      • output/objects
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • bevdet_model_name [default: bevdet_one_lt_d]
      • bevdet_model_path [default: $(var data_path)/tensorrt_bevdet]
  • launch/object_recognition/detection/detector/camera_lidar_detector.launch.xml
      • ns
      • lidar_detection_model_type
      • lidar_detection_model_name
      • use_low_intensity_cluster_filter
      • use_image_segmentation_based_filter
      • number_of_cameras
      • input/camera0/image
      • input/camera0/info
      • input/camera0/rois
      • input/camera1/image
      • input/camera1/info
      • input/camera1/rois
      • input/camera2/image
      • input/camera2/info
      • input/camera2/rois
      • input/camera3/image
      • input/camera3/info
      • input/camera3/rois
      • input/camera4/image
      • input/camera4/info
      • input/camera4/rois
      • input/camera5/image
      • input/camera5/info
      • input/camera5/rois
      • input/camera6/image
      • input/camera6/info
      • input/camera6/rois
      • input/camera7/image
      • input/camera7/info
      • input/camera7/rois
      • input/camera8/image
      • input/camera8/info
      • input/camera8/rois
      • input/concatenation_info
      • segmentation_pointcloud_fusion_camera_ids
      • image_topic_name
      • sync_param_path
      • voxel_grid_based_euclidean_param_path
      • lidar_model_param_path
      • pointpainting_model_path [default: $(var data_path)/image_projection_based_fusion]
      • pointpainting_fusion_common_param_path
      • roi_cluster_fusion_param_path
      • node/pointcloud_container
      • input/pointcloud
      • input/pointcloud_map/pointcloud
      • input/obstacle_segmentation/pointcloud
      • output/ml_detector/objects
      • output/rule_detector/objects
      • camera_2d_detector/model_path
      • camera_2d_detector/label_path
      • camera_2d_detector/color_map_path
      • camera_2d_detector/roi_remap_path
      • camera_2d_detector/roi_to_semseg_remap_path
      • enable_2d_detection [default: false]
  • launch/object_recognition/detection/detector/camera_lidar_irregular_object_detector.launch.xml
      • pipeline_ns
      • node/pointcloud_container
      • output/objects [default: objects]
      • input/concatenation_info
      • input/pointcloud
      • irregular_object_crop_box_filter_node_param_path
      • irregular_object_scan_ground_filter_node_param_path
      • irregular_object_roi_pointcloud_fusion_node_param_path
      • irregular_object_fusion_camera_param_path
      • sync_param_path
  • launch/object_recognition/detection/detector/camera_vru_detector.launch.xml
      • input/camera0/info [default: /sensing/camera/camera0/camera_info]
      • input/camera0/rois [default: /perception/object_recognition/detection/rois0]
      • input/camera1/info [default: /sensing/camera/camera1/camera_info]
      • input/camera1/rois [default: /perception/object_recognition/detection/rois1]
      • input/camera2/info [default: /sensing/camera/camera2/camera_info]
      • input/camera2/rois [default: /perception/object_recognition/detection/rois2]
      • input/camera3/info [default: /sensing/camera/camera3/camera_info]
      • input/camera3/rois [default: /perception/object_recognition/detection/rois3]
      • input/camera4/info [default: /sensing/camera/camera4/camera_info]
      • input/camera4/rois [default: /perception/object_recognition/detection/rois4]
      • input/camera5/info [default: /sensing/camera/camera5/camera_info]
      • input/camera5/rois [default: /perception/object_recognition/detection/rois5]
      • input/camera6/info [default: /sensing/camera/camera6/camera_info]
      • input/camera6/rois [default: /perception/object_recognition/detection/rois6]
      • input/camera7/info [default: /sensing/camera/camera7/camera_info]
      • input/camera7/rois [default: /perception/object_recognition/detection/rois7]
      • output/objects [default: /perception/object_recognition/detection/camera_vru/objects]
      • bbox_object_locator_param_path
      • rois_ids [default: [0, 1]]
  • launch/object_recognition/detection/detector/irregular_object_detector.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • target_container
      • base_frame [default: base_link]
      • use_intra_process [default: true]
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • input/concatenation_info [default: /sensing/lidar/concatenated/pointcloud_info]
      • output/clusters
      • crop_box_filter_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/crop_box_filter.param.yaml]
      • scan_ground_filter_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/scan_ground_filter.param.yaml]
      • roi_pointcloud_fusion_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/roi_pointcloud_fusion.param.yaml]
      • fusion_camera_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/fusion_cameras.param.yaml]
      • sync_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/fusion_common.param.yaml]
  • launch/object_recognition/detection/detector/lidar_dnn_detector.launch.xml
      • lidar_detection_model_type
      • lidar_detection_model_name
      • lidar_model_param_path
      • bevfusion_model_path [default: $(var data_path)/bevfusion]
      • centerpoint_model_path [default: $(var data_path)/lidar_centerpoint]
      • centerpoint_common_param_path [default: $(find-pkg-share autoware_lidar_centerpoint)/config/centerpoint_common.param.yaml]
      • transfusion_model_path [default: $(var data_path)/lidar_transfusion]
      • use_short_range_detection [default: false]
      • lidar_short_range_detection_model_type
      • node/pointcloud_container
      • input/pointcloud
      • output/objects
      • output/short_range_objects
  • launch/object_recognition/detection/detector/lidar_rule_detector.launch.xml
      • ns
      • node/pointcloud_container
      • input/pointcloud_map/pointcloud
      • input/obstacle_segmentation/pointcloud
      • output/cluster_objects
      • output/objects
      • voxel_grid_based_euclidean_param_path
      • use_low_height_cropbox [default: false]
  • launch/object_recognition/detection/detector/lidar_semantic_segmentation_detector.launch.xml
      • euclidean_clustering/param_path
      • input/pointcloud
      • output/objects
  • launch/object_recognition/detection/detector/tracker_based_detector.launch.xml
      • detection_by_tracker_param_path
      • input/clusters
      • input/tracked_objects
      • output/objects
  • launch/object_recognition/detection/filter/object_filter.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • objects_filter_method [default: lanelet_filter]
      • object_recognition_detection_object_lanelet_filter_param_path [default: $(var perception_config_dir)/object_recognition/detection/object_filter/object_lanelet_filter.param.yaml]
      • object_recognition_detection_object_position_filter_param_path [default: $(var perception_config_dir)/object_recognition/detection/object_filter/object_position_filter.param.yaml]
      • input/objects
      • output/objects
  • launch/object_recognition/detection/filter/object_validator.launch.xml
      • objects_validation_method
      • object_recognition_detection_obstacle_pointcloud_based_validator_param_path
      • input/obstacle_pointcloud
      • input/objects
      • output/objects
  • launch/object_recognition/detection/filter/pointcloud_downsample_filter.launch.xml
      • target_container
      • use_intra_process [default: true]
      • input/pointcloud
      • output/pointcloud
      • downsample_voxel_size [default: 0.1]
  • launch/object_recognition/detection/filter/pointcloud_map_filter.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • target_container
      • use_intra_process [default: true]
      • input/pointcloud
      • output/pointcloud
      • output/downsampled/pointcloud [default: /perception/obstacle_segmentation/pointcloud_map_filtered/downsampled/pointcloud]
      • downsample_voxel_size [default: 0.1]
      • voxel_based_compare_map_filter_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/pointcloud_filter/voxel_based_compare_map_filter.param.yaml]
  • launch/object_recognition/detection/filter/radar_filter.launch.xml
      • object_sorter_param_path
      • radar_lanelet_filtering_range_param_path
      • input/radar
      • output/objects
  • launch/object_recognition/detection/merger/camera_lidar_merger.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • object_recognition_detection_object_merger_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/detection/object_merger/data_association_matrix.param.yaml]
      • object_recognition_detection_object_merger_distance_threshold_list_path [default: $(var perception_config_dir)/object_recognition/detection/object_merger/overlapped_judge.param.yaml]
      • ml_camera_lidar_object_association_merger_param_path [default: $(var perception_config_dir)/object_recognition/detection/object_merger/ml_camera_lidar_object_association_merger.param.yaml]
      • roi_detected_object_fusion_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/roi_detected_object_fusion.param.yaml]
      • sync_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/fusion_common.param.yaml]
      • lidar_detection_model_type
      • use_detection_by_tracker
      • use_irregular_object_detector
      • use_roi_detected_object_validator
      • euclidean_clustering/shape_policy
      • use_object_filter
      • objects_filter_method
      • number_of_cameras
      • input/camera0/image
      • input/camera0/info
      • input/camera0/rois
      • input/camera1/image
      • input/camera1/info
      • input/camera1/rois
      • input/camera2/image
      • input/camera2/info
      • input/camera2/rois
      • input/camera3/image
      • input/camera3/info
      • input/camera3/rois
      • input/camera4/image
      • input/camera4/info
      • input/camera4/rois
      • input/camera5/image
      • input/camera5/info
      • input/camera5/rois
      • input/camera6/image
      • input/camera6/info
      • input/camera6/rois
      • input/camera7/image
      • input/camera7/info
      • input/camera7/rois
      • input/camera8/image
      • input/camera8/info
      • input/camera8/rois
      • input/concatenation_info [default: /sensing/lidar/concatenated/pointcloud_info]
      • input/lidar_ml/objects
      • input/lidar_cluster/objects
      • input/irregular_object/objects [default: ]
      • input/detection_by_tracker/objects
      • output/objects [default: objects]
      • alpha_merger_priority_mode [default: 0]
  • launch/object_recognition/object_recognition_camera.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • use_camera_vru_detector [default: false]
      • camera_vru_detector_rois_ids [default: [0]]
      • use_camera_bev_detector [default: true]
      • use_vector_map [default: true]
      • tracker_publish_merged_objects [default: false]
      • input/camera0/image [default: /sensing/camera/camera0/image_rect_color]
      • input/camera0/info [default: /sensing/camera/camera0/camera_info]
      • input/camera0/rois [default: /perception/object_recognition/detection/rois0]
      • input/camera1/image [default: /sensing/camera/camera1/image_rect_color]
      • input/camera1/info [default: /sensing/camera/camera1/camera_info]
      • input/camera1/rois [default: /perception/object_recognition/detection/rois1]
      • input/camera2/image [default: /sensing/camera/camera2/image_rect_color]
      • input/camera2/info [default: /sensing/camera/camera2/camera_info]
      • input/camera2/rois [default: /perception/object_recognition/detection/rois2]
      • input/camera3/image [default: /sensing/camera/camera3/image_rect_color]
      • input/camera3/info [default: /sensing/camera/camera3/camera_info]
      • input/camera3/rois [default: /perception/object_recognition/detection/rois3]
      • input/camera4/image [default: /sensing/camera/camera4/image_rect_color]
      • input/camera4/info [default: /sensing/camera/camera4/camera_info]
      • input/camera4/rois [default: /perception/object_recognition/detection/rois4]
      • input/camera5/image [default: /sensing/camera/camera5/image_rect_color]
      • input/camera5/info [default: /sensing/camera/camera5/camera_info]
      • input/camera5/rois [default: /perception/object_recognition/detection/rois5]
      • input/camera6/image [default: /sensing/camera/camera6/image_rect_color]
      • input/camera6/info [default: /sensing/camera/camera6/camera_info]
      • input/camera6/rois [default: /perception/object_recognition/detection/rois6]
      • input/camera7/image [default: /sensing/camera/camera7/image_rect_color]
      • input/camera7/info [default: /sensing/camera/camera7/camera_info]
      • input/camera7/rois [default: /perception/object_recognition/detection/rois7]
      • output/tracker_merged_objects [default: /perception/object_recognition/detection/objects]
      • output/tracked_objects [default: /perception/object_recognition/tracking/objects]
      • output/predicted_objects [default: /perception/object_recognition/objects]
      • near_range_camera_vru_param_path [default: $(var perception_config_dir)/object_recognition/detection/camera_vru_detection/near_range_camera_vru_detector.param.yaml]
      • object_recognition_tracking_multi_object_tracker_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/data_association_matrix.param.yaml]
      • object_recognition_tracking_multi_object_tracker_input_channels_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/input_channels.param.yaml]
      • object_recognition_tracking_multi_object_tracker_node_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/multi_object_tracker_node.param.yaml]
  • launch/object_recognition/object_recognition_camera_lidar_radar.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • lidar_detection_model [default: centerpoint]
      • use_short_range_detection [default: false]
      • lidar_short_range_detection_model_type [default: centerpoint_short_range]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • bevfusion_model_path [default: $(var data_path)/bevfusion]
      • pointpainting_model_path [default: $(var data_path)/image_projection_based_fusion]
      • camera_2d_detector/model_path [default: $(var data_path)/tensorrt_yolox/yolox-sPlus-opt-pseudoV2-T4-960x960-T4-seg16cls.onnx]
      • camera_2d_detector/label_path [default: $(var data_path)/tensorrt_yolox/label.txt]
      • camera_2d_detector/color_map_path [default: $(var data_path)/tensorrt_yolox/semseg_color_map.csv]
      • camera_2d_detector/roi_remap_path [default: $(var perception_config_dir)/traffic_light_recognition/tensorrt_yolox/roi_label_remap.csv]
      • camera_2d_detector/roi_to_semseg_remap_path [default: $(var perception_config_dir)/traffic_light_recognition/tensorrt_yolox/roi_to_semseg_label_remap.csv]
      • use_semantic_segmentation_ptv3 [default: false]
      • use_object_filter [default: true]
      • use_pointcloud_map [default: true]
      • use_validator [default: false]
      • objects_validation_method [default: obstacle_pointcloud]
      • use_low_height_cropbox [default: false]
      • use_low_intensity_cluster_filter [default: true]
      • use_image_segmentation_based_filter [default: false]
      • use_irregular_object_detector [default: true]
      • use_camera_vru_detector [default: false]
      • camera_vru_detector_rois_ids [default: [0]]
      • use_camera_only_detection [default: false]
      • use_vector_map [default: true]
      • tracker_publish_merged_objects [default: false]
      • number_of_cameras [default: 6]
      • pointcloud_container_name [default: pointcloud_container]
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • input/semantic_segmentation/pointcloud [default: /perception/segmented/pointcloud]
      • input/obstacle_segmentation/pointcloud [default: /perception/obstacle_segmentation/pointcloud]
      • input/camera0/image [default: /sensing/camera/camera0/image_rect_color]
      • input/camera0/info [default: /sensing/camera/camera0/camera_info]
      • input/camera0/rois [default: /perception/object_recognition/detection/rois0]
      • input/camera1/image [default: /sensing/camera/camera1/image_rect_color]
      • input/camera1/info [default: /sensing/camera/camera1/camera_info]
      • input/camera1/rois [default: /perception/object_recognition/detection/rois1]
      • input/camera2/image [default: /sensing/camera/camera2/image_rect_color]
      • input/camera2/info [default: /sensing/camera/camera2/camera_info]
      • input/camera2/rois [default: /perception/object_recognition/detection/rois2]
      • input/camera3/image [default: /sensing/camera/camera3/image_rect_color]
      • input/camera3/info [default: /sensing/camera/camera3/camera_info]
      • input/camera3/rois [default: /perception/object_recognition/detection/rois3]
      • input/camera4/image [default: /sensing/camera/camera4/image_rect_color]
      • input/camera4/info [default: /sensing/camera/camera4/camera_info]
      • input/camera4/rois [default: /perception/object_recognition/detection/rois4]
      • input/camera5/image [default: /sensing/camera/camera5/image_rect_color]
      • input/camera5/info [default: /sensing/camera/camera5/camera_info]
      • input/camera5/rois [default: /perception/object_recognition/detection/rois5]
      • input/camera6/image [default: /sensing/camera/camera6/image_rect_color]
      • input/camera6/info [default: /sensing/camera/camera6/camera_info]
      • input/camera6/rois [default: /perception/object_recognition/detection/rois6]
      • input/camera7/image [default: /sensing/camera/camera7/image_rect_color]
      • input/camera7/info [default: /sensing/camera/camera7/camera_info]
      • input/camera7/rois [default: /perception/object_recognition/detection/rois7]
      • input/camera8/image [default: /sensing/camera/camera8/image_rect_color]
      • input/camera8/info [default: /sensing/camera/camera8/camera_info]
      • input/camera8/rois [default: /perception/object_recognition/detection/rois8]
      • input/concatenation_info [default: /sensing/lidar/concatenated/pointcloud_info]
      • input/radar [default: /sensing/radar/detected_objects]
      • input/camera_only/objects [default: /perception/object_recognition/detection/camera_only/objects]
      • input/camera_only/channel [default: camera_streampetr]
      • image_topic_name [default: image_rect_color]
      • segmentation_pointcloud_fusion_camera_ids [default: [0,2,4]]
      • output/tracker_merged_objects [default: /perception/object_recognition/detection/objects]
      • output/tracked_objects [default: /perception/object_recognition/tracking/objects]
      • output/predicted_objects [default: /perception/object_recognition/objects]
      • sync_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/fusion_common.param.yaml]
      • roi_cluster_fusion_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/roi_cluster_fusion.param.yaml]
      • pointpainting_fusion_common_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/pointpainting_common.param.yaml]
      • irregular_object_crop_box_filter_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/crop_box_filter.param.yaml]
      • irregular_object_scan_ground_filter_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/scan_ground_filter.param.yaml]
      • irregular_object_roi_pointcloud_fusion_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/roi_pointcloud_fusion.param.yaml]
      • irregular_object_fusion_camera_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/fusion_cameras.param.yaml]
      • lidar_model_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model]
      • centerpoint_common_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model/centerpoint_common.param.yaml]
      • voxel_grid_based_euclidean_param_path [default: $(var perception_config_dir)/object_recognition/detection/clustering/voxel_grid_based_euclidean_cluster.param.yaml]
      • radar_lanelet_filtering_range_param_path [default: $(var perception_config_dir)/object_recognition/detection/object_filter/radar_lanelet_filter.param.yaml]
      • object_recognition_detection_object_sorter_radar_param_path [default: $(var perception_config_dir)/object_recognition/detection/radar_filter/detected_object_sorter.param.yaml]
      • near_range_camera_vru_param_path [default: $(var perception_config_dir)/object_recognition/detection/camera_vru_detection/near_range_camera_vru_detector.param.yaml]
      • object_recognition_detection_obstacle_pointcloud_based_validator_param_path [default: $(var perception_config_dir)/object_recognition/detection/detected_object_validation/obstacle_pointcloud_based_validator.param.yaml]
      • object_recognition_tracking_multi_object_tracker_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/data_association_matrix.param.yaml]
      • object_recognition_tracking_multi_object_tracker_input_channels_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/input_channels.param.yaml]
      • object_recognition_tracking_multi_object_tracker_node_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/multi_object_tracker_node.param.yaml]
      • euclidean_clustering/param_path [default: $(var perception_config_dir)/object_recognition/detection/euclidean_cluster/label_based_euclidean_cluster.param.yaml]
  • launch/object_recognition/object_recognition_camera_lidar_radar_serial.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • lidar_detection_model [default: centerpoint]
      • use_short_range_detection [default: false]
      • lidar_short_range_detection_model_type [default: centerpoint_short_range]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • bevfusion_model_path [default: $(var data_path)/bevfusion]
      • pointpainting_model_path [default: $(var data_path)/image_projection_based_fusion]
      • camera_2d_detector/model_path [default: $(var data_path)/tensorrt_yolox/yolox-sPlus-opt-pseudoV2-T4-960x960-T4-seg16cls.onnx]
      • camera_2d_detector/label_path [default: $(var data_path)/tensorrt_yolox/label.txt]
      • camera_2d_detector/color_map_path [default: $(var data_path)/tensorrt_yolox/semseg_color_map.csv]
      • camera_2d_detector/roi_remap_path [default: $(var perception_config_dir)/traffic_light_recognition/tensorrt_yolox/roi_label_remap.csv]
      • camera_2d_detector/roi_to_semseg_remap_path [default: $(var perception_config_dir)/traffic_light_recognition/tensorrt_yolox/roi_to_semseg_label_remap.csv]
      • use_object_filter [default: true]
      • objects_filter_method [default: lanelet_filter]
      • use_pointcloud_map [default: true]
      • use_validator [default: false]
      • objects_validation_method [default: obstacle_pointcloud]
      • use_roi_detected_object_validator [default: true]
      • use_detection_by_tracker [default: true]
      • use_low_height_cropbox [default: false]
      • use_low_intensity_cluster_filter [default: true]
      • use_image_segmentation_based_filter [default: false]
      • use_irregular_object_detector [default: true]
      • ml_camera_lidar_merger_priority_mode [default: 0]
      • use_vector_map [default: true]
      • number_of_cameras [default: 6]
      • pointcloud_container_name [default: pointcloud_container]
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • input/obstacle_segmentation/pointcloud [default: /perception/obstacle_segmentation/pointcloud]
      • input/camera0/image [default: /sensing/camera/camera0/image_rect_color]
      • input/camera0/info [default: /sensing/camera/camera0/camera_info]
      • input/camera0/rois [default: /perception/object_recognition/detection/rois0]
      • input/camera1/image [default: /sensing/camera/camera1/image_rect_color]
      • input/camera1/info [default: /sensing/camera/camera1/camera_info]
      • input/camera1/rois [default: /perception/object_recognition/detection/rois1]
      • input/camera2/image [default: /sensing/camera/camera2/image_rect_color]
      • input/camera2/info [default: /sensing/camera/camera2/camera_info]
      • input/camera2/rois [default: /perception/object_recognition/detection/rois2]
      • input/camera3/image [default: /sensing/camera/camera3/image_rect_color]
      • input/camera3/info [default: /sensing/camera/camera3/camera_info]
      • input/camera3/rois [default: /perception/object_recognition/detection/rois3]
      • input/camera4/image [default: /sensing/camera/camera4/image_rect_color]
      • input/camera4/info [default: /sensing/camera/camera4/camera_info]
      • input/camera4/rois [default: /perception/object_recognition/detection/rois4]
      • input/camera5/image [default: /sensing/camera/camera5/image_rect_color]
      • input/camera5/info [default: /sensing/camera/camera5/camera_info]
      • input/camera5/rois [default: /perception/object_recognition/detection/rois5]
      • input/camera6/image [default: /sensing/camera/camera6/image_rect_color]
      • input/camera6/info [default: /sensing/camera/camera6/camera_info]
      • input/camera6/rois [default: /perception/object_recognition/detection/rois6]
      • input/camera7/image [default: /sensing/camera/camera7/image_rect_color]
      • input/camera7/info [default: /sensing/camera/camera7/camera_info]
      • input/camera7/rois [default: /perception/object_recognition/detection/rois7]
      • input/camera8/image [default: /sensing/camera/camera8/image_rect_color]
      • input/camera8/info [default: /sensing/camera/camera8/camera_info]
      • input/camera8/rois [default: /perception/object_recognition/detection/rois8]
      • input/concatenation_info [default: /sensing/lidar/concatenated/pointcloud_info]
      • input/radar/tracked_objects [default: /sensing/radar/tracked_objects]
      • image_topic_name [default: image_rect_color]
      • segmentation_pointcloud_fusion_camera_ids [default: [0,2,4]]
      • output/tracked_objects [default: /perception/object_recognition/tracking/objects]
      • output/predicted_objects [default: /perception/object_recognition/objects]
      • sync_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/fusion_common.param.yaml]
      • roi_cluster_fusion_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/roi_cluster_fusion.param.yaml]
      • pointpainting_fusion_common_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/pointpainting_common.param.yaml]
      • irregular_object_crop_box_filter_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/crop_box_filter.param.yaml]
      • irregular_object_scan_ground_filter_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/scan_ground_filter.param.yaml]
      • irregular_object_roi_pointcloud_fusion_node_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/roi_pointcloud_fusion.param.yaml]
      • irregular_object_fusion_camera_param_path [default: $(var perception_config_dir)/object_recognition/detection/irregular_object_detection/fusion_cameras.param.yaml]
      • lidar_model_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model]
      • centerpoint_common_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model/centerpoint_common.param.yaml]
      • voxel_grid_based_euclidean_param_path [default: $(var perception_config_dir)/object_recognition/detection/clustering/voxel_grid_based_euclidean_cluster.param.yaml]
      • detection_by_tracker_param_path [default: $(var perception_config_dir)/object_recognition/detection/detection_by_tracker/detection_by_tracker.param.yaml]
      • object_recognition_detection_obstacle_pointcloud_based_validator_param_path [default: $(var perception_config_dir)/object_recognition/detection/detected_object_validation/obstacle_pointcloud_based_validator.param.yaml]
      • object_recognition_detection_object_merger_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/detection/object_merger/data_association_matrix.param.yaml]
      • object_recognition_detection_object_merger_distance_threshold_list_path [default: $(var perception_config_dir)/object_recognition/detection/object_merger/overlapped_judge.param.yaml]
      • ml_camera_lidar_object_association_merger_param_path [default: $(var perception_config_dir)/object_recognition/detection/object_merger/ml_camera_lidar_object_association_merger.param.yaml]
      • roi_detected_object_fusion_param_path [default: $(var perception_config_dir)/object_recognition/detection/image_projection_based_fusion/roi_detected_object_fusion.param.yaml]
      • object_recognition_tracking_multi_object_tracker_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/data_association_matrix.param.yaml]
      • object_recognition_tracking_multi_object_tracker_input_channels_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/input_channels.param.yaml]
      • object_recognition_tracking_multi_object_tracker_node_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/multi_object_tracker_node.param.yaml]
      • object_recognition_tracking_radar_tracked_object_sorter_param_path [default: $(var perception_config_dir)/object_recognition/tracking/radar_object_tracker/tracked_object_sorter.param.yaml]
      • object_recognition_tracking_radar_tracked_object_lanelet_filter_param_path [default: $(var perception_config_dir)/object_recognition/tracking/radar_object_tracker/tracked_object_lanelet_filter.param.yaml]
      • object_recognition_tracking_object_merger_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/tracking/tracking_object_merger/data_association_matrix.param.yaml]
      • object_recognition_tracking_object_merger_node_param_path [default: $(var perception_config_dir)/object_recognition/tracking/tracking_object_merger/decorative_tracker_merger.param.yaml]
      • euclidean_clustering/shape_policy [default: 0]
  • launch/object_recognition/object_recognition_lidar.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • lidar_detection_model [default: centerpoint]
      • use_short_range_detection [default: false]
      • lidar_short_range_detection_model_type [default: centerpoint_short_range]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • bevfusion_model_path [default: $(var data_path)/bevfusion]
      • use_pointcloud_map [default: true]
      • use_validator [default: false]
      • objects_validation_method [default: obstacle_pointcloud]
      • use_object_filter [default: true]
      • objects_filter_method [default: lanelet_filter]
      • use_low_height_cropbox [default: false]
      • use_vector_map [default: true]
      • tracker_publish_merged_objects [default: false]
      • pointcloud_container_name [default: pointcloud_container]
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • input/obstacle_segmentation/pointcloud [default: /perception/obstacle_segmentation/pointcloud]
      • output/tracker_merged_objects [default: /perception/object_recognition/detection/objects]
      • output/tracked_objects [default: /perception/object_recognition/tracking/objects]
      • output/predicted_objects [default: /perception/object_recognition/objects]
      • lidar_model_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model]
      • centerpoint_common_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model/centerpoint_common.param.yaml]
      • voxel_grid_based_euclidean_param_path [default: $(var perception_config_dir)/object_recognition/detection/clustering/voxel_grid_based_euclidean_cluster.param.yaml]
      • object_recognition_detection_obstacle_pointcloud_based_validator_param_path [default: $(var perception_config_dir)/object_recognition/detection/detected_object_validation/obstacle_pointcloud_based_validator.param.yaml]
      • object_recognition_tracking_multi_object_tracker_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/data_association_matrix.param.yaml]
      • object_recognition_tracking_multi_object_tracker_input_channels_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/input_channels.param.yaml]
      • object_recognition_tracking_multi_object_tracker_node_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/multi_object_tracker_node.param.yaml]
  • launch/object_recognition/object_recognition_lidar_ptv3.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • lidar_detection_model [default: centerpoint]
      • use_short_range_detection [default: false]
      • lidar_short_range_detection_model_type [default: centerpoint_short_range]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • bevfusion_model_path [default: $(var data_path)/bevfusion]
      • use_vector_map [default: true]
      • tracker_publish_merged_objects [default: false]
      • pointcloud_container_name [default: pointcloud_container]
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • input/semantic_segmentation/pointcloud [default: /perception/segmented/pointcloud]
      • output/tracker_merged_objects [default: /perception/object_recognition/detection/objects]
      • output/tracked_objects [default: /perception/object_recognition/tracking/objects]
      • output/predicted_objects [default: /perception/object_recognition/objects]
      • lidar_model_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model]
      • centerpoint_common_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model/centerpoint_common.param.yaml]
      • object_recognition_tracking_multi_object_tracker_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/data_association_matrix.param.yaml]
      • object_recognition_tracking_multi_object_tracker_input_channels_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/input_channels.param.yaml]
      • object_recognition_tracking_multi_object_tracker_node_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/multi_object_tracker_node.param.yaml]
      • euclidean_clustering/param_path [default: $(var perception_config_dir)/object_recognition/detection/euclidean_cluster/label_based_euclidean_cluster.param.yaml]
  • launch/object_recognition/object_recognition_lidar_radar.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • lidar_detection_model [default: centerpoint]
      • use_short_range_detection [default: false]
      • lidar_short_range_detection_model_type [default: centerpoint_short_range]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • bevfusion_model_path [default: $(var data_path)/bevfusion]
      • use_low_height_cropbox [default: false]
      • use_vector_map [default: true]
      • tracker_publish_merged_objects [default: false]
      • pointcloud_container_name [default: pointcloud_container]
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • input/semantic_segmentation/pointcloud [default: /perception/segmented/pointcloud]
      • input/radar [default: /sensing/radar/detected_objects]
      • output/tracker_merged_objects [default: /perception/object_recognition/detection/objects]
      • output/tracked_objects [default: /perception/object_recognition/tracking/objects]
      • output/predicted_objects [default: /perception/object_recognition/objects]
      • lidar_model_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model]
      • centerpoint_common_param_path [default: $(var perception_config_dir)/object_recognition/detection/lidar_model/centerpoint_common.param.yaml]
      • voxel_grid_based_euclidean_param_path [default: $(var perception_config_dir)/object_recognition/detection/clustering/voxel_grid_based_euclidean_cluster.param.yaml]
      • radar_lanelet_filtering_range_param_path [default: $(var perception_config_dir)/object_recognition/detection/object_filter/radar_lanelet_filter.param.yaml]
      • object_recognition_detection_object_sorter_radar_param_path [default: $(var perception_config_dir)/object_recognition/detection/radar_filter/detected_object_sorter.param.yaml]
      • object_recognition_tracking_multi_object_tracker_data_association_matrix_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/data_association_matrix.param.yaml]
      • object_recognition_tracking_multi_object_tracker_input_channels_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/input_channels.param.yaml]
      • object_recognition_tracking_multi_object_tracker_node_param_path [default: $(var perception_config_dir)/object_recognition/tracking/multi_object_tracker/multi_object_tracker_node.param.yaml]
      • euclidean_clustering/param_path [default: $(var perception_config_dir)/object_recognition/detection/euclidean_cluster/label_based_euclidean_cluster.param.yaml]
  • launch/object_recognition/prediction/prediction.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • use_vector_map [default: false]
      • prediction_model_type [default: map_based]
      • input/objects [default: /perception/object_recognition/tracking/objects]
      • output/objects [default: /perception/object_recognition/objects]
      • object_recognition_prediction_map_based_prediction_param_path [default: $(var perception_config_dir)/object_recognition/prediction/map_based_prediction.param.yaml]
  • launch/obstacle_segmentation/ground_segmentation_additional_lidars.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • target_container
      • lidar_name
      • base_frame [default: base_link]
      • use_intra_process [default: true]
      • input/main/pointcloud
      • input/lidar/pointcloud [default: /sensing/lidar/$(var lidar_name)/pointcloud]
      • output/pointcloud
      • output/pointcloud_info [default: $(var output/pointcloud)_info]
      • crop_box_margin_max_z [default: 2.5]
      • crop_box_margin_min_z [default: -2.5]
      • crop_box_filter_node_param_path [default: $(var perception_config_dir)/obstacle_segmentation/ground_segmentation/crop_box_filter.param.yaml]
      • scan_ground_filter_node_param_path [default: $(var perception_config_dir)/obstacle_segmentation/ground_segmentation/scan_ground_filter.param.yaml]
  • launch/obstacle_segmentation/ground_segmentation_ml.launch.xml
      • pointcloud_container_name [default: pointcloud_container]
      • use_pointcloud_container [default: true]
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • output/pointcloud [default: /perception/obstacle_segmentation/pointcloud]
      • output/pointcloud/segmentation [default: /perception/segmented/pointcloud]
      • output/pointcloud/visualization [default: /perception/segmented/pointcloud/visualization]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • model_variant [default: ptv3]
      • model_path [default: $(var data_path)/ptv3]
  • launch/obstacle_segmentation/ground_segmentation_ransac.launch.xml
      • target_container
      • base_frame [default: base_link]
      • use_intra_process [default: true]
      • input_topics
      • input/main/pointcloud
      • output/pointcloud
      • output/pointcloud_info [default: $(var output/pointcloud)_info]
      • detection_area_crop_box_filter_node_param_path
      • ransac_ground_filter_node_param_path [default: $(find-pkg-share autoware_ground_segmentation)/config/ransac_ground_filter.param.yaml]
  • launch/obstacle_segmentation/ground_segmentation_rule_based.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • target_container
      • base_frame [default: base_link]
      • use_intra_process [default: true]
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • output/pointcloud [default: /perception/obstacle_segmentation/pointcloud]
      • crop_box_margin_max_z [default: 2.5]
      • crop_box_margin_min_z [default: -2.5]
      • crop_box_filter_node_param_path [default: $(var perception_config_dir)/obstacle_segmentation/ground_segmentation/crop_box_filter.param.yaml]
      • scan_ground_filter_node_param_path [default: $(var perception_config_dir)/obstacle_segmentation/ground_segmentation/scan_ground_filter.param.yaml]
  • launch/obstacle_segmentation/ground_segmentation_rule_based_cuda.launch.xml
      • target_container
      • input/pointcloud [default: /sensing/lidar/concatenated/pointcloud]
      • output/pointcloud [default: /perception/obstacle_segmentation/pointcloud]
      • output/ground_pointcloud [default: /perception/obstacle_segmentation/ground_pointcloud]
      • cuda_scan_ground_segmentation_filter_node_param_path [default: $(find-pkg-share autoware_ground_segmentation_cuda)/config/cuda_scan_ground_segmentation_filter.param.yaml]
  • launch/occupancy_grid_map/probabilistic_occupancy_grid_map.launch.xml
      • perception_config_pkg [default: autoware_perception_config]
      • input/obstacle_pointcloud [default: concatenated/pointcloud]
      • input/raw_pointcloud [default: no_ground/oneshot/pointcloud]
      • output [default: /perception/occupancy_grid_map/map]
      • use_intra_process [default: false]
      • use_multithread [default: false]
      • pointcloud_container_name [default: pointcloud_container]
      • occupancy_grid_map_method
      • occupancy_grid_map_param_path [default: $(var perception_config_dir)/occupancy_grid_map/$(var occupancy_grid_map_method).param.yaml]
      • occupancy_grid_map_updater
      • occupancy_grid_map_updater_param_path [default: $(var perception_config_dir)/occupancy_grid_map/$(var occupancy_grid_map_updater)_updater.param.yaml]
      • input_obstacle_pointcloud [default: false]
      • input_obstacle_and_raw_pointcloud [default: true]
      • use_pointcloud_container [default: true]
  • launch/traffic_light_recognition/traffic_light_detector_fine.launch.xml
    • Fine-detection chain for one traffic light camera: map based detector proposes rough ROIs, the fine detector refines them, classifiers label them.
      • perception_config_pkg [default: autoware_perception_config]
      • camera_namespace
      • enable_image_decompressor [default: true]
      • input/vector_map [default: /map/vector_map]
      • input/route [default: /planning/mission_planning/route]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • map_based_detector_param_path [default: $(find-pkg-share autoware_traffic_light_map_based_detector)/config/traffic_light_map_based_detector.param.yaml]
      • fine_detector_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_fine_detector/traffic_light_fine_detector.param.yaml]
      • car_classifier_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_classifier/car_traffic_light_classifier.param.yaml]
      • pedestrian_classifier_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_classifier/pedestrian_traffic_light_classifier.param.yaml]
      • lamp_recognizer_ml_param_path [default: $(var data_path)/traffic_light_classifier/lamp_recognizer_ml.param.yaml]
      • roi_visualizer_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_visualization/traffic_light_roi_visualizer.param.yaml]
      • fine_detector_model_path [default: $(var data_path)/traffic_light_fine_detector/tlr_car_ped_yolox_s_batch_6.onnx]
      • fine_detector_label_path [default: $(var data_path)/traffic_light_fine_detector/tlr_labels.txt]
      • car_classifier_model_path [default: $(var data_path)/traffic_light_classifier/traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • car_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels.txt]
      • pedestrian_classifier_model_path [default: $(var data_path)/traffic_light_classifier/ped_traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • pedestrian_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels_ped.txt]
      • car_classifier_type [default: 1]
      • pedestrian_classifier_type [default: 1]
  • launch/traffic_light_recognition/traffic_light_detector_rough.launch.xml
    • Rough-detection chain for one traffic light camera: map based detector ROIs feed the classifiers directly, without a high accuracy detector.
      • perception_config_pkg [default: autoware_perception_config]
      • camera_namespace
      • enable_image_decompressor [default: true]
      • input/vector_map [default: /map/vector_map]
      • input/route [default: /planning/mission_planning/route]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • map_based_detector_param_path [default: $(find-pkg-share autoware_traffic_light_map_based_detector)/config/traffic_light_map_based_detector.param.yaml]
      • car_classifier_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_classifier/car_traffic_light_classifier.param.yaml]
      • pedestrian_classifier_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_classifier/pedestrian_traffic_light_classifier.param.yaml]
      • lamp_recognizer_ml_param_path [default: $(var data_path)/traffic_light_classifier/lamp_recognizer_ml.param.yaml]
      • roi_visualizer_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_visualization/traffic_light_roi_visualizer.param.yaml]
      • car_classifier_model_path [default: $(var data_path)/traffic_light_classifier/traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • car_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels.txt]
      • pedestrian_classifier_model_path [default: $(var data_path)/traffic_light_classifier/ped_traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • pedestrian_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels_ped.txt]
      • car_classifier_type [default: 1]
      • pedestrian_classifier_type [default: 1]
  • launch/traffic_light_recognition/traffic_light_detector_whole_image.launch.xml
    • Whole-image detection chain for one traffic light camera: yolox detects over the full image, the selector matches detections to map based ROIs, classifiers label them.
      • perception_config_pkg [default: autoware_perception_config]
      • camera_namespace
      • enable_image_decompressor [default: true]
      • input/vector_map [default: /map/vector_map]
      • input/route [default: /planning/mission_planning/route]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • map_based_detector_param_path [default: $(find-pkg-share autoware_traffic_light_map_based_detector)/config/traffic_light_map_based_detector.param.yaml]
      • whole_image_detector_param_path [default: $(var perception_config_dir)/traffic_light_recognition/tensorrt_yolox/yolox_traffic_light_detector.param.yaml]
      • car_classifier_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_classifier/car_traffic_light_classifier.param.yaml]
      • pedestrian_classifier_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_classifier/pedestrian_traffic_light_classifier.param.yaml]
      • lamp_recognizer_ml_param_path [default: $(var data_path)/traffic_light_classifier/lamp_recognizer_ml.param.yaml]
      • roi_visualizer_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_visualization/traffic_light_roi_visualizer.param.yaml]
      • whole_image_detector_model_path [default: $(var data_path)/tensorrt_yolox/yolox_s_car_ped_tl_detector_960_960_batch_1.onnx]
      • whole_image_detector_label_path [default: $(var data_path)/tensorrt_yolox/car_ped_tl_detector_labels.txt]
      • car_classifier_model_path [default: $(var data_path)/traffic_light_classifier/traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • car_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels.txt]
      • pedestrian_classifier_model_path [default: $(var data_path)/traffic_light_classifier/ped_traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • pedestrian_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels_ped.txt]
      • car_classifier_type [default: 1]
      • pedestrian_classifier_type [default: 1]
  • launch/traffic_light_recognition/traffic_light_fusion.launch.xml
    • Post-process shared by all traffic light cameras: multi camera fusion, V2X arbitration, pedestrian signal estimation and map visualization.
      • perception_config_pkg [default: autoware_perception_config]
      • camera_namespaces [default: [camera6, camera7]]
      • input/vector_map [default: /map/vector_map]
      • external/traffic_signals [default: /perception/traffic_light_recognition/external/traffic_signals]
      • internal/traffic_signals [default: /perception/traffic_light_recognition/internal/traffic_signals]
      • judged/traffic_signals [default: /perception/traffic_light_recognition/judged/traffic_signals]
      • output/traffic_signals [default: /perception/traffic_light_recognition/traffic_signals]
      • multi_camera_fusion_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_multi_camera_fusion/traffic_light_multi_camera_fusion.param.yaml]
      • arbiter_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_arbiter/traffic_light_arbiter.param.yaml]
      • crosswalk_estimator_param_path [default: $(var perception_config_dir)/traffic_light_recognition/crosswalk_traffic_light_estimator/crosswalk_traffic_light_estimator.param.yaml]
  • launch/traffic_light_recognition/traffic_light_occlusion_predictor.launch.xml
    • Occlusion check for one traffic light camera: merges the car and pedestrian signals into the per-camera output, masking signals hidden from the camera.
      • perception_config_pkg [default: autoware_perception_config]
      • camera_namespace
      • input/cloud [default: /sensing/lidar/top/pointcloud_raw_ex]
      • input/vector_map [default: /map/vector_map]
      • occlusion_predictor_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_occlusion_predictor/traffic_light_occlusion_predictor.param.yaml]
  • launch/traffic_light_recognition/traffic_light_recognition_fine.launch.xml
    • Traffic light recognition, fine detection mode: per-camera fine detection with occlusion check, then shared fusion. Two cameras as the representative set; add or remove per-camera blocks to match the vehicle.
      • perception_config_pkg [default: autoware_perception_config]
      • input/vector_map [default: /map/vector_map]
      • input/route [default: /planning/mission_planning/route]
      • input/pointcloud [default: /sensing/lidar/top/pointcloud_raw_ex]
      • external/traffic_signals [default: /perception/traffic_light_recognition/external/traffic_signals]
      • output/traffic_signals [default: /perception/traffic_light_recognition/traffic_signals]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • fine_detector_model_path [default: $(var data_path)/traffic_light_fine_detector/tlr_car_ped_yolox_s_batch_6.onnx]
      • fine_detector_label_path [default: $(var data_path)/traffic_light_fine_detector/tlr_labels.txt]
      • lamp_recognizer_ml_param_path [default: $(var data_path)/traffic_light_classifier/lamp_recognizer_ml.param.yaml]
      • car_classifier_model_path [default: $(var data_path)/traffic_light_classifier/traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • car_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels.txt]
      • pedestrian_classifier_model_path [default: $(var data_path)/traffic_light_classifier/ped_traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • pedestrian_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels_ped.txt]
      • car_classifier_type [default: 1]
      • pedestrian_classifier_type [default: 1]
      • base_map_based_detector_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_map_based_detector/base_traffic_light_map_based_detector.param.yaml]
      • extra_map_based_detector_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_map_based_detector/extra_traffic_light_map_based_detector.param.yaml]
  • launch/traffic_light_recognition/traffic_light_recognition_merge.launch.xml
    • Traffic light recognition, merge mode: detection and classification run on other ECUs; this side runs the per-camera occlusion check and the shared fusion. Two cameras as the representative set; add or remove per-camera blocks to match the vehicle.
      • input/vector_map [default: /map/vector_map]
      • input/pointcloud [default: /sensing/lidar/top/pointcloud_raw_ex]
      • external/traffic_signals [default: /perception/traffic_light_recognition/external/traffic_signals]
      • output/traffic_signals [default: /perception/traffic_light_recognition/traffic_signals]
  • launch/traffic_light_recognition/traffic_light_recognition_rough.launch.xml
    • Traffic light recognition, rough detection mode: per-camera map based detection with occlusion check, then shared fusion. One camera as the representative set; add or remove per-camera blocks to match the vehicle.
      • perception_config_pkg [default: autoware_perception_config]
      • input/vector_map [default: /map/vector_map]
      • input/route [default: /planning/mission_planning/route]
      • input/pointcloud [default: /sensing/lidar/top/pointcloud_raw_ex]
      • external/traffic_signals [default: /perception/traffic_light_recognition/external/traffic_signals]
      • output/traffic_signals [default: /perception/traffic_light_recognition/traffic_signals]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • lamp_recognizer_ml_param_path [default: $(var data_path)/traffic_light_classifier/lamp_recognizer_ml.param.yaml]
      • car_classifier_model_path [default: $(var data_path)/traffic_light_classifier/traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • car_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels.txt]
      • pedestrian_classifier_model_path [default: $(var data_path)/traffic_light_classifier/ped_traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • pedestrian_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels_ped.txt]
      • car_classifier_type [default: 1]
      • pedestrian_classifier_type [default: 1]
      • base_map_based_detector_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_map_based_detector/base_traffic_light_map_based_detector.param.yaml]
  • launch/traffic_light_recognition/traffic_light_recognition_whole_image.launch.xml
    • Traffic light recognition, whole-image detection mode: per-camera yolox detection, then shared fusion. Two cameras as the representative set; add or remove per-camera blocks to match the vehicle.
      • perception_config_pkg [default: autoware_perception_config]
      • input/vector_map [default: /map/vector_map]
      • input/route [default: /planning/mission_planning/route]
      • external/traffic_signals [default: /perception/traffic_light_recognition/external/traffic_signals]
      • output/traffic_signals [default: /perception/traffic_light_recognition/traffic_signals]
      • data_path [default: $(env HOME)/autoware_data/ml_models]
      • whole_image_detector_model_path [default: $(var data_path)/tensorrt_yolox/yolox_s_car_ped_tl_detector_960_960_batch_1.onnx]
      • whole_image_detector_label_path [default: $(var data_path)/tensorrt_yolox/car_ped_tl_detector_labels.txt]
      • lamp_recognizer_ml_param_path [default: $(var data_path)/traffic_light_classifier/lamp_recognizer_ml.param.yaml]
      • car_classifier_model_path [default: $(var data_path)/traffic_light_classifier/traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • car_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels.txt]
      • pedestrian_classifier_model_path [default: $(var data_path)/traffic_light_classifier/ped_traffic_light_classifier_mobilenetv2_batch_6.onnx]
      • pedestrian_classifier_label_path [default: $(var data_path)/traffic_light_classifier/lamp_labels_ped.txt]
      • car_classifier_type [default: 1]
      • pedestrian_classifier_type [default: 1]
      • base_map_based_detector_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_map_based_detector/base_traffic_light_map_based_detector.param.yaml]
      • extra_map_based_detector_param_path [default: $(var perception_config_dir)/traffic_light_recognition/traffic_light_map_based_detector/extra_traffic_light_map_based_detector.param.yaml]

Messages

No message files found.

Services

No service files found

Plugins

No plugins found.

Recent questions tagged autoware_perception_launch at Robotics Stack Exchange