Skip to content

A repository for autonomous racing written in ROS2, designed for the F1TENTH platform. Code can run both in the physical car, as well as simulation with custom launch files.

License

Notifications You must be signed in to change notification settings

CL2-UWaterloo/f1tenth_ws

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

f1tenth_ws

This a repository that contains ready-to-run autonomous racing packages for the F1TENTH on ROS2 (>= Foxy). It can be directly be deployed on the physical car. We've also included launch and config files for the simulation environment, which uses slightly different topics for odometry.

Below is a demo of the car running the code from this repository in the E7 building at the University of Waterloo at a top speed of ~25km/h, record on March 30th 2023.

00108_4.mp4

Software Stack Overview

Our current software stack consists of

  • slam_toolbox for mapping (reference the following slides for running it on the physical car)
  • Particle Filter for localization $\rightarrow$ src/particle_filter
  • Pure Pursuit for waypoint following (planning + control) $\rightarrow$ src/pure_pursuit
  • Stanley Avoidance for a waypoint following algorithm that includes dynamic obstacle avoidance (slightly slower), includes an improved controller (using Stanley method) $\rightarrow$ src/stanley_avoidance.

Racing lines are generated through the Cl2-UWaterloo/Raceline-Optimization repository.

Other algorithms that are not used, but are in this repository include

  • Waypoint Generator for manually generating waypoints in simulation $\rightarrow$ src/waypoint_generator (this has been replaced with a script that automatically generates optimal racelines given a map generated by slam_toolbox)
  • A PID controller for staying at a constant distance to the wall $\rightarrow$ src/wall_follow
  • Scan matching $\rightarrow$ src/scan_matching (To be completed)
  • gap_follow $\rightarrow$ src/gap_follow (To be completed)

High-Level Usage Guide

These are the high level steps followed to get the F1TENTH driving in a new location, with accompanying notes:

  1. Run SLAM on the physical car to generate a map with slam_toolbox
  2. Clean up map in Photoshop, and generate a racing line using the Cl2-UWaterloo/Raceline-Optimization repository.
  3. Store the racing lines under src/pure_pursuit/racelines/
  4. Run particle_filter with the new map to localize the car properly
  5. Run pure_pursuit or rrt (for obstacle avoidance) to follow the racing line. Make sure to incrementally increase the velocity_profile inside the config.yaml file.

Getting Started

If you are a complete beginner and want to build an autonomous RC car, you might want to first follow the F1TENTH Course, where they teach you ROS2 and the intuition behind the algorithms in this repository. The official documentation contains information on setting up the hardware and software for the F1TENTH car.

You can also consult https://stevengong.co/notes/F1TENTH for quick links to different resources.

Once you have access to an F1TENTH vehicle, you can consult our accompanying notes for the commands used to deploy the code in this repository: https://stevengong.co/notes/F1TENTH-Field-Usage. Note that they are mainly written for our own personal reference, so the paths will be different in your setup. You'll notice that we clone the repository on our local computer, and upload the code over SSH to the car.

Alternatively, you can try this code out in simulation without a physical car (see below).

Getting Started in Simulation

If you want to try out the ROS nodes in simulation, you can follow the steps below.We've tested these steps on a Ubuntu 20.04 native machine using Docker.

First, we will need to clone the simulation repository by running

git clone https://github.com/f1tenth/f1tenth_gym_ros

Then, build the docker image by running:

cd f1tenth_gym_ros
sudo docker build -t f1tenth_gym_ros -f Dockerfile .

You then need to mount the f1tenth_ws repository into the docker container. You will need to change the path in the command below:

cd <INSERT_PATH_HERE>/f1tenth_ws/src
sudo rocker --nvidia --x11 --volume .:/sim_ws/src -- f1tenth_gym_ros

You will then be entered the docker container. The f1tenth_ws code is mounted under /sim_ws/src/

Now, we need to load/change the default map in the simulation. In the container, modify the map_path parameter inside the sim.yaml file (located at /sim_ws/install/f1tenth_gym_ros/share/f1tenth_gym_ros/config/sim.yaml) to the path of the map you desire.

For example,

map_path: '/sim_ws/src/particle_filter/maps/e7_floor5'

Run the following commands from the /sim_ws directory:

source /opt/ros/foxy/setup.bash
source install/local_setup.bash
ros2 launch f1tenth_gym_ros gym_bridge_launch.py

Other tip: when your vehicle ends up in really bad spots, you can just use the "2D Pose Estimate" button inside rviz and select where you want to respawn the vehicle. Quite convenient!

To run the ROS2 nodes from this repository (ex: pure_pursuit), you need to run these commands from /sim_ws/src/f1tenth_gym_ros in another window (tip: use tmux):

colcon build
. install/setup.bash
ros2 launch pure_pursuit sim_pure_pursuit_launch.py

If you don't know what the above commands do, you probably should get familiar with ROS2 first.

Next Steps

A non-exhaustive list of things we want to do in the future, include

  • Running SLAM + pure pursuit on the fly, without having to do the offline computation. Something like this would be incredible
  • Trying out MPC, something like this is awesome!

About Us

The Control, Learning and Logic (CL2) group at the University of Waterloo works on research that aims to develop methods for reliable decision-making of autonomous systems in the wild, led by professor Yash Vardhan Pant. Current members working on the F1TENTH is composed of Steven Gong, Oluwatofolafun Damilola Opeoluwa-Calebs, and Soham Lakhi.

About

A repository for autonomous racing written in ROS2, designed for the F1TENTH platform. Code can run both in the physical car, as well as simulation with custom launch files.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published