Package Summary
Tags | No category tags. |
Version | 0.1.0 |
License | Apache License 2.0 |
Build type | AMENT_CMAKE |
Use | RECOMMENDED |
Repository Summary
Description | |
Checkout URI | https://github.com/ieiauto/autodrrt.git |
VCS Type | git |
VCS Version | main |
Last Updated | 2025-05-30 |
Dev Status | UNKNOWN |
Released | UNRELEASED |
Tags | No category tags. |
Contributing |
Help Wanted (-)
Good First Issues (-) Pull Requests to Review (-) |
Package Description
Additional Links
Maintainers
- Taiki Tanaka
- Tomoya Kimura
- Shumpei Wakabayashi
- Fumiya Watanabe
- Takamasa Horibe
- Satoshi Ota
- Takayuki Murooka
Authors
- Satoshi Tanaka
obstacle_collision_checker
Purpose
obstacle_collision_checker
is a module to check obstacle collision for predicted trajectory and publish diagnostic errors if collision is found.
Inner-workings / Algorithms
Flow chart
@startuml
skinparam monochrome true
title obstacle collision checker : update
start
:calculate braking distance;
:resampling trajectory;
note right
to reduce calculation cost
end note
:filter point cloud by trajectory;
:create vehicle foot prints;
:create vehicle passing area;
partition will_collide {
while (has next ego vehicle foot print) is (yes)
:found collision with obstacle foot print;
if (has collision with obstacle) then (yes)
:set diag to ERROR;
stop
endif
end while (no)
:set diag to OK;
stop
}
@enduml
Algorithms
Check data
Check that obstacle_collision_checker
receives no ground pointcloud, predicted_trajectory, reference trajectory, and current velocity data.
Diagnostic update
If any collision is found on predicted path, this module sets ERROR
level as diagnostic status else sets OK
.
Inputs / Outputs
Input
Name | Type | Description |
---|---|---|
~/input/trajectory |
autoware_auto_planning_msgs::msg::Trajectory |
Reference trajectory |
~/input/trajectory |
autoware_auto_planning_msgs::msg::Trajectory |
Predicted trajectory |
/perception/obstacle_segmentation/pointcloud |
sensor_msgs::msg::PointCloud2 |
Pointcloud of obstacles which the ego-vehicle should stop or avoid |
/tf |
tf2_msgs::msg::TFMessage |
TF |
/tf_static |
tf2_msgs::msg::TFMessage |
TF static |
Output
Name | Type | Description |
---|---|---|
~/debug/marker |
visualization_msgs::msg::MarkerArray |
Marker for visualization |
Parameters
Name | Type | Description | Default value |
---|---|---|---|
delay_time |
double |
Delay time of vehicle [s] | 0.3 |
footprint_margin |
double |
Foot print margin [m] | 0.0 |
max_deceleration |
double |
Max deceleration for ego vehicle to stop [m/s^2] | 2.0 |
resample_interval |
double |
Interval for resampling trajectory [m] | 0.3 |
search_radius |
double |
Search distance from trajectory to point cloud [m] | 5.0 |
Assumptions / Known limits
To perform proper collision check, it is necessary to get probably predicted trajectory and obstacle pointclouds without noise.
Package Dependencies
System Dependencies
Dependant Packages
Launch files
- launch/obstacle_collision_checker.launch.xml
-
- input/lanelet_map_bin [default: /map/vector_map]
- input/obstacle_pointcloud [default: /perception/obstacle_segmentation/pointcloud]
- input/reference_trajectory [default: /planning/scenario_planning/trajectory]
- input/predicted_trajectory [default: /control/trajectory_follower/predicted_trajectory]
- input/odometry [default: /localization/kinematic_state]
- config_file [default: $(find-pkg-share obstacle_collision_checker)/config/obstacle_collision_checker.param.yaml]
- vehicle_info_param_file [default: $(find-pkg-share vehicle_info_util)/config/vehicle_info.param.yaml]
Messages
Services
Plugins
Recent questions tagged obstacle_collision_checker at Robotics Stack Exchange
Package Summary
Tags | No category tags. |
Version | 0.1.0 |
License | Apache License 2.0 |
Build type | AMENT_CMAKE |
Use | RECOMMENDED |
Repository Summary
Description | |
Checkout URI | https://github.com/ieiauto/autodrrt.git |
VCS Type | git |
VCS Version | main |
Last Updated | 2025-05-30 |
Dev Status | UNKNOWN |
Released | UNRELEASED |
Tags | No category tags. |
Contributing |
Help Wanted (-)
Good First Issues (-) Pull Requests to Review (-) |
Package Description
Additional Links
Maintainers
- Taiki Tanaka
- Tomoya Kimura
- Shumpei Wakabayashi
- Fumiya Watanabe
- Takamasa Horibe
- Satoshi Ota
- Takayuki Murooka
Authors
- Satoshi Tanaka
obstacle_collision_checker
Purpose
obstacle_collision_checker
is a module to check obstacle collision for predicted trajectory and publish diagnostic errors if collision is found.
Inner-workings / Algorithms
Flow chart
@startuml
skinparam monochrome true
title obstacle collision checker : update
start
:calculate braking distance;
:resampling trajectory;
note right
to reduce calculation cost
end note
:filter point cloud by trajectory;
:create vehicle foot prints;
:create vehicle passing area;
partition will_collide {
while (has next ego vehicle foot print) is (yes)
:found collision with obstacle foot print;
if (has collision with obstacle) then (yes)
:set diag to ERROR;
stop
endif
end while (no)
:set diag to OK;
stop
}
@enduml
Algorithms
Check data
Check that obstacle_collision_checker
receives no ground pointcloud, predicted_trajectory, reference trajectory, and current velocity data.
Diagnostic update
If any collision is found on predicted path, this module sets ERROR
level as diagnostic status else sets OK
.
Inputs / Outputs
Input
Name | Type | Description |
---|---|---|
~/input/trajectory |
autoware_auto_planning_msgs::msg::Trajectory |
Reference trajectory |
~/input/trajectory |
autoware_auto_planning_msgs::msg::Trajectory |
Predicted trajectory |
/perception/obstacle_segmentation/pointcloud |
sensor_msgs::msg::PointCloud2 |
Pointcloud of obstacles which the ego-vehicle should stop or avoid |
/tf |
tf2_msgs::msg::TFMessage |
TF |
/tf_static |
tf2_msgs::msg::TFMessage |
TF static |
Output
Name | Type | Description |
---|---|---|
~/debug/marker |
visualization_msgs::msg::MarkerArray |
Marker for visualization |
Parameters
Name | Type | Description | Default value |
---|---|---|---|
delay_time |
double |
Delay time of vehicle [s] | 0.3 |
footprint_margin |
double |
Foot print margin [m] | 0.0 |
max_deceleration |
double |
Max deceleration for ego vehicle to stop [m/s^2] | 2.0 |
resample_interval |
double |
Interval for resampling trajectory [m] | 0.3 |
search_radius |
double |
Search distance from trajectory to point cloud [m] | 5.0 |
Assumptions / Known limits
To perform proper collision check, it is necessary to get probably predicted trajectory and obstacle pointclouds without noise.
Package Dependencies
System Dependencies
Dependant Packages
Launch files
- launch/obstacle_collision_checker.launch.xml
-
- input/lanelet_map_bin [default: /map/vector_map]
- input/obstacle_pointcloud [default: /perception/obstacle_segmentation/pointcloud]
- input/reference_trajectory [default: /planning/scenario_planning/trajectory]
- input/predicted_trajectory [default: /control/trajectory_follower/predicted_trajectory]
- input/odometry [default: /localization/kinematic_state]
- config_file [default: $(find-pkg-share obstacle_collision_checker)/config/obstacle_collision_checker.param.yaml]
- vehicle_info_param_file [default: $(find-pkg-share vehicle_info_util)/config/vehicle_info.param.yaml]
Messages
Services
Plugins
Recent questions tagged obstacle_collision_checker at Robotics Stack Exchange
Package Summary
Tags | No category tags. |
Version | 0.1.0 |
License | Apache License 2.0 |
Build type | AMENT_CMAKE |
Use | RECOMMENDED |
Repository Summary
Description | |
Checkout URI | https://github.com/ieiauto/autodrrt.git |
VCS Type | git |
VCS Version | main |
Last Updated | 2025-05-30 |
Dev Status | UNKNOWN |
Released | UNRELEASED |
Tags | No category tags. |
Contributing |
Help Wanted (-)
Good First Issues (-) Pull Requests to Review (-) |
Package Description
Additional Links
Maintainers
- Taiki Tanaka
- Tomoya Kimura
- Shumpei Wakabayashi
- Fumiya Watanabe
- Takamasa Horibe
- Satoshi Ota
- Takayuki Murooka
Authors
- Satoshi Tanaka
obstacle_collision_checker
Purpose
obstacle_collision_checker
is a module to check obstacle collision for predicted trajectory and publish diagnostic errors if collision is found.
Inner-workings / Algorithms
Flow chart
@startuml
skinparam monochrome true
title obstacle collision checker : update
start
:calculate braking distance;
:resampling trajectory;
note right
to reduce calculation cost
end note
:filter point cloud by trajectory;
:create vehicle foot prints;
:create vehicle passing area;
partition will_collide {
while (has next ego vehicle foot print) is (yes)
:found collision with obstacle foot print;
if (has collision with obstacle) then (yes)
:set diag to ERROR;
stop
endif
end while (no)
:set diag to OK;
stop
}
@enduml
Algorithms
Check data
Check that obstacle_collision_checker
receives no ground pointcloud, predicted_trajectory, reference trajectory, and current velocity data.
Diagnostic update
If any collision is found on predicted path, this module sets ERROR
level as diagnostic status else sets OK
.
Inputs / Outputs
Input
Name | Type | Description |
---|---|---|
~/input/trajectory |
autoware_auto_planning_msgs::msg::Trajectory |
Reference trajectory |
~/input/trajectory |
autoware_auto_planning_msgs::msg::Trajectory |
Predicted trajectory |
/perception/obstacle_segmentation/pointcloud |
sensor_msgs::msg::PointCloud2 |
Pointcloud of obstacles which the ego-vehicle should stop or avoid |
/tf |
tf2_msgs::msg::TFMessage |
TF |
/tf_static |
tf2_msgs::msg::TFMessage |
TF static |
Output
Name | Type | Description |
---|---|---|
~/debug/marker |
visualization_msgs::msg::MarkerArray |
Marker for visualization |
Parameters
Name | Type | Description | Default value |
---|---|---|---|
delay_time |
double |
Delay time of vehicle [s] | 0.3 |
footprint_margin |
double |
Foot print margin [m] | 0.0 |
max_deceleration |
double |
Max deceleration for ego vehicle to stop [m/s^2] | 2.0 |
resample_interval |
double |
Interval for resampling trajectory [m] | 0.3 |
search_radius |
double |
Search distance from trajectory to point cloud [m] | 5.0 |
Assumptions / Known limits
To perform proper collision check, it is necessary to get probably predicted trajectory and obstacle pointclouds without noise.
Package Dependencies
System Dependencies
Dependant Packages
Launch files
- launch/obstacle_collision_checker.launch.xml
-
- input/lanelet_map_bin [default: /map/vector_map]
- input/obstacle_pointcloud [default: /perception/obstacle_segmentation/pointcloud]
- input/reference_trajectory [default: /planning/scenario_planning/trajectory]
- input/predicted_trajectory [default: /control/trajectory_follower/predicted_trajectory]
- input/odometry [default: /localization/kinematic_state]
- config_file [default: $(find-pkg-share obstacle_collision_checker)/config/obstacle_collision_checker.param.yaml]
- vehicle_info_param_file [default: $(find-pkg-share vehicle_info_util)/config/vehicle_info.param.yaml]
Messages
Services
Plugins
Recent questions tagged obstacle_collision_checker at Robotics Stack Exchange
Package Summary
Tags | No category tags. |
Version | 0.1.0 |
License | Apache License 2.0 |
Build type | AMENT_CMAKE |
Use | RECOMMENDED |
Repository Summary
Description | |
Checkout URI | https://github.com/ieiauto/autodrrt.git |
VCS Type | git |
VCS Version | main |
Last Updated | 2025-05-30 |
Dev Status | UNKNOWN |
Released | UNRELEASED |
Tags | No category tags. |
Contributing |
Help Wanted (-)
Good First Issues (-) Pull Requests to Review (-) |
Package Description
Additional Links
Maintainers
- Taiki Tanaka
- Tomoya Kimura
- Shumpei Wakabayashi
- Fumiya Watanabe
- Takamasa Horibe
- Satoshi Ota
- Takayuki Murooka
Authors
- Satoshi Tanaka
obstacle_collision_checker
Purpose
obstacle_collision_checker
is a module to check obstacle collision for predicted trajectory and publish diagnostic errors if collision is found.
Inner-workings / Algorithms
Flow chart
@startuml
skinparam monochrome true
title obstacle collision checker : update
start
:calculate braking distance;
:resampling trajectory;
note right
to reduce calculation cost
end note
:filter point cloud by trajectory;
:create vehicle foot prints;
:create vehicle passing area;
partition will_collide {
while (has next ego vehicle foot print) is (yes)
:found collision with obstacle foot print;
if (has collision with obstacle) then (yes)
:set diag to ERROR;
stop
endif
end while (no)
:set diag to OK;
stop
}
@enduml
Algorithms
Check data
Check that obstacle_collision_checker
receives no ground pointcloud, predicted_trajectory, reference trajectory, and current velocity data.
Diagnostic update
If any collision is found on predicted path, this module sets ERROR
level as diagnostic status else sets OK
.
Inputs / Outputs
Input
Name | Type | Description |
---|---|---|
~/input/trajectory |
autoware_auto_planning_msgs::msg::Trajectory |
Reference trajectory |
~/input/trajectory |
autoware_auto_planning_msgs::msg::Trajectory |
Predicted trajectory |
/perception/obstacle_segmentation/pointcloud |
sensor_msgs::msg::PointCloud2 |
Pointcloud of obstacles which the ego-vehicle should stop or avoid |
/tf |
tf2_msgs::msg::TFMessage |
TF |
/tf_static |
tf2_msgs::msg::TFMessage |
TF static |
Output
Name | Type | Description |
---|---|---|
~/debug/marker |
visualization_msgs::msg::MarkerArray |
Marker for visualization |
Parameters
Name | Type | Description | Default value |
---|---|---|---|
delay_time |
double |
Delay time of vehicle [s] | 0.3 |
footprint_margin |
double |
Foot print margin [m] | 0.0 |
max_deceleration |
double |
Max deceleration for ego vehicle to stop [m/s^2] | 2.0 |
resample_interval |
double |
Interval for resampling trajectory [m] | 0.3 |
search_radius |
double |
Search distance from trajectory to point cloud [m] | 5.0 |
Assumptions / Known limits
To perform proper collision check, it is necessary to get probably predicted trajectory and obstacle pointclouds without noise.
Package Dependencies
System Dependencies
Dependant Packages
Launch files
- launch/obstacle_collision_checker.launch.xml
-
- input/lanelet_map_bin [default: /map/vector_map]
- input/obstacle_pointcloud [default: /perception/obstacle_segmentation/pointcloud]
- input/reference_trajectory [default: /planning/scenario_planning/trajectory]
- input/predicted_trajectory [default: /control/trajectory_follower/predicted_trajectory]
- input/odometry [default: /localization/kinematic_state]
- config_file [default: $(find-pkg-share obstacle_collision_checker)/config/obstacle_collision_checker.param.yaml]
- vehicle_info_param_file [default: $(find-pkg-share vehicle_info_util)/config/vehicle_info.param.yaml]
Messages
Services
Plugins
Recent questions tagged obstacle_collision_checker at Robotics Stack Exchange
Package Summary
Tags | No category tags. |
Version | 0.1.0 |
License | Apache License 2.0 |
Build type | AMENT_CMAKE |
Use | RECOMMENDED |
Repository Summary
Description | |
Checkout URI | https://github.com/ieiauto/autodrrt.git |
VCS Type | git |
VCS Version | main |
Last Updated | 2025-05-30 |
Dev Status | UNKNOWN |
Released | UNRELEASED |
Tags | No category tags. |
Contributing |
Help Wanted (-)
Good First Issues (-) Pull Requests to Review (-) |
Package Description
Additional Links
Maintainers
- Taiki Tanaka
- Tomoya Kimura
- Shumpei Wakabayashi
- Fumiya Watanabe
- Takamasa Horibe
- Satoshi Ota
- Takayuki Murooka
Authors
- Satoshi Tanaka
obstacle_collision_checker
Purpose
obstacle_collision_checker
is a module to check obstacle collision for predicted trajectory and publish diagnostic errors if collision is found.
Inner-workings / Algorithms
Flow chart
@startuml
skinparam monochrome true
title obstacle collision checker : update
start
:calculate braking distance;
:resampling trajectory;
note right
to reduce calculation cost
end note
:filter point cloud by trajectory;
:create vehicle foot prints;
:create vehicle passing area;
partition will_collide {
while (has next ego vehicle foot print) is (yes)
:found collision with obstacle foot print;
if (has collision with obstacle) then (yes)
:set diag to ERROR;
stop
endif
end while (no)
:set diag to OK;
stop
}
@enduml
Algorithms
Check data
Check that obstacle_collision_checker
receives no ground pointcloud, predicted_trajectory, reference trajectory, and current velocity data.
Diagnostic update
If any collision is found on predicted path, this module sets ERROR
level as diagnostic status else sets OK
.
Inputs / Outputs
Input
Name | Type | Description |
---|---|---|
~/input/trajectory |
autoware_auto_planning_msgs::msg::Trajectory |
Reference trajectory |
~/input/trajectory |
autoware_auto_planning_msgs::msg::Trajectory |
Predicted trajectory |
/perception/obstacle_segmentation/pointcloud |
sensor_msgs::msg::PointCloud2 |
Pointcloud of obstacles which the ego-vehicle should stop or avoid |
/tf |
tf2_msgs::msg::TFMessage |
TF |
/tf_static |
tf2_msgs::msg::TFMessage |
TF static |
Output
Name | Type | Description |
---|---|---|
~/debug/marker |
visualization_msgs::msg::MarkerArray |
Marker for visualization |
Parameters
Name | Type | Description | Default value |
---|---|---|---|
delay_time |
double |
Delay time of vehicle [s] | 0.3 |
footprint_margin |
double |
Foot print margin [m] | 0.0 |
max_deceleration |
double |
Max deceleration for ego vehicle to stop [m/s^2] | 2.0 |
resample_interval |
double |
Interval for resampling trajectory [m] | 0.3 |
search_radius |
double |
Search distance from trajectory to point cloud [m] | 5.0 |
Assumptions / Known limits
To perform proper collision check, it is necessary to get probably predicted trajectory and obstacle pointclouds without noise.
Package Dependencies
System Dependencies
Dependant Packages
Launch files
- launch/obstacle_collision_checker.launch.xml
-
- input/lanelet_map_bin [default: /map/vector_map]
- input/obstacle_pointcloud [default: /perception/obstacle_segmentation/pointcloud]
- input/reference_trajectory [default: /planning/scenario_planning/trajectory]
- input/predicted_trajectory [default: /control/trajectory_follower/predicted_trajectory]
- input/odometry [default: /localization/kinematic_state]
- config_file [default: $(find-pkg-share obstacle_collision_checker)/config/obstacle_collision_checker.param.yaml]
- vehicle_info_param_file [default: $(find-pkg-share vehicle_info_util)/config/vehicle_info.param.yaml]
Messages
Services
Plugins
Recent questions tagged obstacle_collision_checker at Robotics Stack Exchange
Package Summary
Tags | No category tags. |
Version | 0.1.0 |
License | Apache License 2.0 |
Build type | AMENT_CMAKE |
Use | RECOMMENDED |
Repository Summary
Description | |
Checkout URI | https://github.com/ieiauto/autodrrt.git |
VCS Type | git |
VCS Version | main |
Last Updated | 2025-05-30 |
Dev Status | UNKNOWN |
Released | UNRELEASED |
Tags | No category tags. |
Contributing |
Help Wanted (-)
Good First Issues (-) Pull Requests to Review (-) |
Package Description
Additional Links
Maintainers
- Taiki Tanaka
- Tomoya Kimura
- Shumpei Wakabayashi
- Fumiya Watanabe
- Takamasa Horibe
- Satoshi Ota
- Takayuki Murooka
Authors
- Satoshi Tanaka
obstacle_collision_checker
Purpose
obstacle_collision_checker
is a module to check obstacle collision for predicted trajectory and publish diagnostic errors if collision is found.
Inner-workings / Algorithms
Flow chart
@startuml
skinparam monochrome true
title obstacle collision checker : update
start
:calculate braking distance;
:resampling trajectory;
note right
to reduce calculation cost
end note
:filter point cloud by trajectory;
:create vehicle foot prints;
:create vehicle passing area;
partition will_collide {
while (has next ego vehicle foot print) is (yes)
:found collision with obstacle foot print;
if (has collision with obstacle) then (yes)
:set diag to ERROR;
stop
endif
end while (no)
:set diag to OK;
stop
}
@enduml
Algorithms
Check data
Check that obstacle_collision_checker
receives no ground pointcloud, predicted_trajectory, reference trajectory, and current velocity data.
Diagnostic update
If any collision is found on predicted path, this module sets ERROR
level as diagnostic status else sets OK
.
Inputs / Outputs
Input
Name | Type | Description |
---|---|---|
~/input/trajectory |
autoware_auto_planning_msgs::msg::Trajectory |
Reference trajectory |
~/input/trajectory |
autoware_auto_planning_msgs::msg::Trajectory |
Predicted trajectory |
/perception/obstacle_segmentation/pointcloud |
sensor_msgs::msg::PointCloud2 |
Pointcloud of obstacles which the ego-vehicle should stop or avoid |
/tf |
tf2_msgs::msg::TFMessage |
TF |
/tf_static |
tf2_msgs::msg::TFMessage |
TF static |
Output
Name | Type | Description |
---|---|---|
~/debug/marker |
visualization_msgs::msg::MarkerArray |
Marker for visualization |
Parameters
Name | Type | Description | Default value |
---|---|---|---|
delay_time |
double |
Delay time of vehicle [s] | 0.3 |
footprint_margin |
double |
Foot print margin [m] | 0.0 |
max_deceleration |
double |
Max deceleration for ego vehicle to stop [m/s^2] | 2.0 |
resample_interval |
double |
Interval for resampling trajectory [m] | 0.3 |
search_radius |
double |
Search distance from trajectory to point cloud [m] | 5.0 |
Assumptions / Known limits
To perform proper collision check, it is necessary to get probably predicted trajectory and obstacle pointclouds without noise.
Package Dependencies
System Dependencies
Dependant Packages
Launch files
- launch/obstacle_collision_checker.launch.xml
-
- input/lanelet_map_bin [default: /map/vector_map]
- input/obstacle_pointcloud [default: /perception/obstacle_segmentation/pointcloud]
- input/reference_trajectory [default: /planning/scenario_planning/trajectory]
- input/predicted_trajectory [default: /control/trajectory_follower/predicted_trajectory]
- input/odometry [default: /localization/kinematic_state]
- config_file [default: $(find-pkg-share obstacle_collision_checker)/config/obstacle_collision_checker.param.yaml]
- vehicle_info_param_file [default: $(find-pkg-share vehicle_info_util)/config/vehicle_info.param.yaml]
Messages
Services
Plugins
Recent questions tagged obstacle_collision_checker at Robotics Stack Exchange
Package Summary
Tags | No category tags. |
Version | 0.1.0 |
License | Apache License 2.0 |
Build type | AMENT_CMAKE |
Use | RECOMMENDED |
Repository Summary
Description | |
Checkout URI | https://github.com/ieiauto/autodrrt.git |
VCS Type | git |
VCS Version | main |
Last Updated | 2025-05-30 |
Dev Status | UNKNOWN |
Released | UNRELEASED |
Tags | No category tags. |
Contributing |
Help Wanted (-)
Good First Issues (-) Pull Requests to Review (-) |
Package Description
Additional Links
Maintainers
- Taiki Tanaka
- Tomoya Kimura
- Shumpei Wakabayashi
- Fumiya Watanabe
- Takamasa Horibe
- Satoshi Ota
- Takayuki Murooka
Authors
- Satoshi Tanaka
obstacle_collision_checker
Purpose
obstacle_collision_checker
is a module to check obstacle collision for predicted trajectory and publish diagnostic errors if collision is found.
Inner-workings / Algorithms
Flow chart
@startuml
skinparam monochrome true
title obstacle collision checker : update
start
:calculate braking distance;
:resampling trajectory;
note right
to reduce calculation cost
end note
:filter point cloud by trajectory;
:create vehicle foot prints;
:create vehicle passing area;
partition will_collide {
while (has next ego vehicle foot print) is (yes)
:found collision with obstacle foot print;
if (has collision with obstacle) then (yes)
:set diag to ERROR;
stop
endif
end while (no)
:set diag to OK;
stop
}
@enduml
Algorithms
Check data
Check that obstacle_collision_checker
receives no ground pointcloud, predicted_trajectory, reference trajectory, and current velocity data.
Diagnostic update
If any collision is found on predicted path, this module sets ERROR
level as diagnostic status else sets OK
.
Inputs / Outputs
Input
Name | Type | Description |
---|---|---|
~/input/trajectory |
autoware_auto_planning_msgs::msg::Trajectory |
Reference trajectory |
~/input/trajectory |
autoware_auto_planning_msgs::msg::Trajectory |
Predicted trajectory |
/perception/obstacle_segmentation/pointcloud |
sensor_msgs::msg::PointCloud2 |
Pointcloud of obstacles which the ego-vehicle should stop or avoid |
/tf |
tf2_msgs::msg::TFMessage |
TF |
/tf_static |
tf2_msgs::msg::TFMessage |
TF static |
Output
Name | Type | Description |
---|---|---|
~/debug/marker |
visualization_msgs::msg::MarkerArray |
Marker for visualization |
Parameters
Name | Type | Description | Default value |
---|---|---|---|
delay_time |
double |
Delay time of vehicle [s] | 0.3 |
footprint_margin |
double |
Foot print margin [m] | 0.0 |
max_deceleration |
double |
Max deceleration for ego vehicle to stop [m/s^2] | 2.0 |
resample_interval |
double |
Interval for resampling trajectory [m] | 0.3 |
search_radius |
double |
Search distance from trajectory to point cloud [m] | 5.0 |
Assumptions / Known limits
To perform proper collision check, it is necessary to get probably predicted trajectory and obstacle pointclouds without noise.
Package Dependencies
System Dependencies
Dependant Packages
Launch files
- launch/obstacle_collision_checker.launch.xml
-
- input/lanelet_map_bin [default: /map/vector_map]
- input/obstacle_pointcloud [default: /perception/obstacle_segmentation/pointcloud]
- input/reference_trajectory [default: /planning/scenario_planning/trajectory]
- input/predicted_trajectory [default: /control/trajectory_follower/predicted_trajectory]
- input/odometry [default: /localization/kinematic_state]
- config_file [default: $(find-pkg-share obstacle_collision_checker)/config/obstacle_collision_checker.param.yaml]
- vehicle_info_param_file [default: $(find-pkg-share vehicle_info_util)/config/vehicle_info.param.yaml]
Messages
Services
Plugins
Recent questions tagged obstacle_collision_checker at Robotics Stack Exchange
Package Summary
Tags | No category tags. |
Version | 0.1.0 |
License | Apache License 2.0 |
Build type | AMENT_CMAKE |
Use | RECOMMENDED |
Repository Summary
Description | |
Checkout URI | https://github.com/ieiauto/autodrrt.git |
VCS Type | git |
VCS Version | main |
Last Updated | 2025-05-30 |
Dev Status | UNKNOWN |
Released | UNRELEASED |
Tags | No category tags. |
Contributing |
Help Wanted (-)
Good First Issues (-) Pull Requests to Review (-) |
Package Description
Additional Links
Maintainers
- Taiki Tanaka
- Tomoya Kimura
- Shumpei Wakabayashi
- Fumiya Watanabe
- Takamasa Horibe
- Satoshi Ota
- Takayuki Murooka
Authors
- Satoshi Tanaka
obstacle_collision_checker
Purpose
obstacle_collision_checker
is a module to check obstacle collision for predicted trajectory and publish diagnostic errors if collision is found.
Inner-workings / Algorithms
Flow chart
@startuml
skinparam monochrome true
title obstacle collision checker : update
start
:calculate braking distance;
:resampling trajectory;
note right
to reduce calculation cost
end note
:filter point cloud by trajectory;
:create vehicle foot prints;
:create vehicle passing area;
partition will_collide {
while (has next ego vehicle foot print) is (yes)
:found collision with obstacle foot print;
if (has collision with obstacle) then (yes)
:set diag to ERROR;
stop
endif
end while (no)
:set diag to OK;
stop
}
@enduml
Algorithms
Check data
Check that obstacle_collision_checker
receives no ground pointcloud, predicted_trajectory, reference trajectory, and current velocity data.
Diagnostic update
If any collision is found on predicted path, this module sets ERROR
level as diagnostic status else sets OK
.
Inputs / Outputs
Input
Name | Type | Description |
---|---|---|
~/input/trajectory |
autoware_auto_planning_msgs::msg::Trajectory |
Reference trajectory |
~/input/trajectory |
autoware_auto_planning_msgs::msg::Trajectory |
Predicted trajectory |
/perception/obstacle_segmentation/pointcloud |
sensor_msgs::msg::PointCloud2 |
Pointcloud of obstacles which the ego-vehicle should stop or avoid |
/tf |
tf2_msgs::msg::TFMessage |
TF |
/tf_static |
tf2_msgs::msg::TFMessage |
TF static |
Output
Name | Type | Description |
---|---|---|
~/debug/marker |
visualization_msgs::msg::MarkerArray |
Marker for visualization |
Parameters
Name | Type | Description | Default value |
---|---|---|---|
delay_time |
double |
Delay time of vehicle [s] | 0.3 |
footprint_margin |
double |
Foot print margin [m] | 0.0 |
max_deceleration |
double |
Max deceleration for ego vehicle to stop [m/s^2] | 2.0 |
resample_interval |
double |
Interval for resampling trajectory [m] | 0.3 |
search_radius |
double |
Search distance from trajectory to point cloud [m] | 5.0 |
Assumptions / Known limits
To perform proper collision check, it is necessary to get probably predicted trajectory and obstacle pointclouds without noise.
Package Dependencies
System Dependencies
Dependant Packages
Launch files
- launch/obstacle_collision_checker.launch.xml
-
- input/lanelet_map_bin [default: /map/vector_map]
- input/obstacle_pointcloud [default: /perception/obstacle_segmentation/pointcloud]
- input/reference_trajectory [default: /planning/scenario_planning/trajectory]
- input/predicted_trajectory [default: /control/trajectory_follower/predicted_trajectory]
- input/odometry [default: /localization/kinematic_state]
- config_file [default: $(find-pkg-share obstacle_collision_checker)/config/obstacle_collision_checker.param.yaml]
- vehicle_info_param_file [default: $(find-pkg-share vehicle_info_util)/config/vehicle_info.param.yaml]
Messages
Services
Plugins
Recent questions tagged obstacle_collision_checker at Robotics Stack Exchange
Package Summary
Tags | No category tags. |
Version | 0.1.0 |
License | Apache License 2.0 |
Build type | AMENT_CMAKE |
Use | RECOMMENDED |
Repository Summary
Description | |
Checkout URI | https://github.com/ieiauto/autodrrt.git |
VCS Type | git |
VCS Version | main |
Last Updated | 2025-05-30 |
Dev Status | UNKNOWN |
Released | UNRELEASED |
Tags | No category tags. |
Contributing |
Help Wanted (-)
Good First Issues (-) Pull Requests to Review (-) |
Package Description
Additional Links
Maintainers
- Taiki Tanaka
- Tomoya Kimura
- Shumpei Wakabayashi
- Fumiya Watanabe
- Takamasa Horibe
- Satoshi Ota
- Takayuki Murooka
Authors
- Satoshi Tanaka
obstacle_collision_checker
Purpose
obstacle_collision_checker
is a module to check obstacle collision for predicted trajectory and publish diagnostic errors if collision is found.
Inner-workings / Algorithms
Flow chart
@startuml
skinparam monochrome true
title obstacle collision checker : update
start
:calculate braking distance;
:resampling trajectory;
note right
to reduce calculation cost
end note
:filter point cloud by trajectory;
:create vehicle foot prints;
:create vehicle passing area;
partition will_collide {
while (has next ego vehicle foot print) is (yes)
:found collision with obstacle foot print;
if (has collision with obstacle) then (yes)
:set diag to ERROR;
stop
endif
end while (no)
:set diag to OK;
stop
}
@enduml
Algorithms
Check data
Check that obstacle_collision_checker
receives no ground pointcloud, predicted_trajectory, reference trajectory, and current velocity data.
Diagnostic update
If any collision is found on predicted path, this module sets ERROR
level as diagnostic status else sets OK
.
Inputs / Outputs
Input
Name | Type | Description |
---|---|---|
~/input/trajectory |
autoware_auto_planning_msgs::msg::Trajectory |
Reference trajectory |
~/input/trajectory |
autoware_auto_planning_msgs::msg::Trajectory |
Predicted trajectory |
/perception/obstacle_segmentation/pointcloud |
sensor_msgs::msg::PointCloud2 |
Pointcloud of obstacles which the ego-vehicle should stop or avoid |
/tf |
tf2_msgs::msg::TFMessage |
TF |
/tf_static |
tf2_msgs::msg::TFMessage |
TF static |
Output
Name | Type | Description |
---|---|---|
~/debug/marker |
visualization_msgs::msg::MarkerArray |
Marker for visualization |
Parameters
Name | Type | Description | Default value |
---|---|---|---|
delay_time |
double |
Delay time of vehicle [s] | 0.3 |
footprint_margin |
double |
Foot print margin [m] | 0.0 |
max_deceleration |
double |
Max deceleration for ego vehicle to stop [m/s^2] | 2.0 |
resample_interval |
double |
Interval for resampling trajectory [m] | 0.3 |
search_radius |
double |
Search distance from trajectory to point cloud [m] | 5.0 |
Assumptions / Known limits
To perform proper collision check, it is necessary to get probably predicted trajectory and obstacle pointclouds without noise.
Package Dependencies
System Dependencies
Dependant Packages
Launch files
- launch/obstacle_collision_checker.launch.xml
-
- input/lanelet_map_bin [default: /map/vector_map]
- input/obstacle_pointcloud [default: /perception/obstacle_segmentation/pointcloud]
- input/reference_trajectory [default: /planning/scenario_planning/trajectory]
- input/predicted_trajectory [default: /control/trajectory_follower/predicted_trajectory]
- input/odometry [default: /localization/kinematic_state]
- config_file [default: $(find-pkg-share obstacle_collision_checker)/config/obstacle_collision_checker.param.yaml]
- vehicle_info_param_file [default: $(find-pkg-share vehicle_info_util)/config/vehicle_info.param.yaml]