Package Summary
Tags | No category tags. |
Version | 0.0.0 |
License | MIT |
Build type | CATKIN |
Use | RECOMMENDED |
Repository Summary
Description | F1Tenth Simulation Code: Platooning, Computer Vision, Reinforcement Learning, Path Planning |
Checkout URI | https://github.com/pmusau17/platooning-f1tenth.git |
VCS Type | git |
VCS Version | noetic-port |
Last Updated | 2022-05-20 |
Dev Status | UNKNOWN |
CI status | No Continuous Integration |
Released | UNRELEASED |
Tags | No category tags. |
Contributing |
Help Wanted (0)
Good First Issues (0) Pull Requests to Review (0) |
Package Description
Additional Links
Maintainers
- Patrick Musau
Authors
Path Tracking: Pure Pursuit
Implementation of the pure pursuit algorithm proposed by R. Craig Coulter in the paper “Implementation of the Pure Pursuit Path Tracking Algorithm”.
Description
The pure pursuit algorithm is a path tracking algorithm that seeks to determine the speed and steering angle neccessary to control a robot to follow a specified path. A path consists of a set of points representing the positional coordinates of a particular course. Intuitively, what the algorithm does is obtain a goal-point, then aim the vehicle towards that point. The algorithm has a single tunable parameter called the lookahead distance and that dictates how the goal point is selected.
Waypoints
To collect waypoints you must specify two things: The world name and the name of the car launched in the simulation. The available world names are listed below. As an example to collect waypoints in the track_porto world run the following and then drive the car manually with the keyboard:
Terminal 1:
roslaunch race f1_tenth_devel.launch world_name:=track_porto enable_keyboard:=true
Terminal 2:
rosrun pure_pursuit waypoint_logger_pure_pursuit.py /racecar
Available Worlds
- track_barca: Large racetrack usually used to evaluate efficacy of algorithms developed on racetracks
- track_levine: Racetrack inspired by the Levine building at UPenn
- track_porto: Simple racetrack with small and wide turns
- racecar_walker: Beautiful track with multiple paths,
- racecar_cones: Empty world with safety cones spaced evenly.
- racecar_tunnel: World with car located in a tunnel with high walls
- racecar_parking_1: Racecar world for you to test your paralle parking skills
- lab_track2: Simple racetrack modeled after laboratory space at Vanderbilt University
Running Experiments
rosrun pure_pursuit pure_pursuit.py /racecar track_porto_26780.csv
Visualizing Experiments
rosrun pure_pursuit visualize_markers.py track_porto_26780.csv
Wiki Tutorials
Dependant Packages
Launch files
- launch/pure_pursuit.launch
-
- world_name [default: track_porto]
- csv_filename [default: track_porto_26780.csv]
- car_name [default: racecar]
- launch/collect_waypoints.launch
-
- world_name [default: track_porto]
- car_name [default: racecar]
- launch/pure_pursuit_particle_filter.launch
-
- map_file [default: $(find a_stars_pure_pursuit)/maps/levine.yaml]
- scan_topic [default: /scan]
- odometry_topic [default: /vesc/odom]
- angle_step [default: 18]
- max_particles [default: 4000]
- squash_factor [default: 2.2]
- viz [default: 1]