F1TENTH gym environment ROS2 communication bridge

This is a containerized ROS communication bridge for the F1TENTH gym environment that turns it into a simulation in ROS2.

Videos

Visit this link to view some experimental simulation videos.

Installation

Native on Ubuntu 20.04

Install the following dependencies:

• ROS 2 Follow the instructions here to install ROS 2 Foxy.
• F1TENTH Gym

  git clone https://github.com/f1tenth/f1tenth_gym
  cd f1tenth_gym && pip3 install -e .

Installing the simulation:

• Create a workspace: cd $HOME && mkdir -p sim_ws/src
• Clone the repo into the workspace:

  cd $HOME/sim_ws/src
  git clone https://github.com/f1tenth/f1tenth_gym_ros_ntu

• Update correct parameter for path to map file: Go to sim.yaml https://github.com/f1tenth/f1tenth_gym_ros_ntu/blob/main/config/sim.yaml in your cloned repo, change the map_path parameter to point to the correct location. It should be '<your_home_dir>/sim_ws/src/f1tenth_gym_ros_ntu/maps/levine'
• Install dependencies with rosdep:

  source /opt/ros/foxy/setup.bash
  cd ..
  rosdep install -i --from-path src --rosdistro foxy -y

• Build the workspace: colcon build

Configuring the simulation

• The configuration file for the simulation is at f1tenth_gym_ros_ntu/config/sim.yaml.
• Topic names and namespaces can be configured but is recommended to leave uncahnged.
• The map can be changed via the map_path parameter. You'll have to use the full path to the map file in the container. It is assumed that the image file and the yaml file for the map are in the same directory with the same name.
• There are 6 maps (levine, levine_closed, Montreal, Sao Paulo, Austin, Spielberg). levine_closed is a closed version of levine for min time raceline optimization. Set the map name correctly in sim.yaml

Running the Simulation

1) Simple Pure Pursuit Controller

Terminal 1: Rviz Window

cd sim_ws
source install/setup.bash
ros2 launch f1tenth_gym_ros gym_bridge_launch.py

Terminal 2: Run Pure Pursuit Controller

cd sim_ws
source install/setup.bash
ros2 run f1tenth_gym_ros simple_pp

Do change the variable map_name to exact map name as defined in sim.yaml. The car should now track the waypoint.

2) Residual RL (SAC) Controller - No Obstacles - levine_closed map

Ensure that the map_name and map_path are correct and match with the csv file name in tracks folder

Terminal 1: Rviz Window

cd sim_ws
source install/setup.bash
ros2 launch f1tenth_gym_ros gym_bridge_launch.py

The script to run RL is test.py in RL folder. Make sure the variables exp_no, algo, logs_type, obstacle_type match with the trained weights in logs/ folder.

Terminal 2: Run Pure Pursuit Controller

cd sim_ws/src/f1tenth_gym_ros
python3 RL/test.py

If everything goes correct, you should see the car moving with the residual RL controller.

3) Residual RL (SAC) Controller - Static Obstacles - levine_closed_static_obstacle map

Do the following changes in sim.yaml

# map parameters
map_name: 'levine_closed_static_obstacle'
map_path: '/sim_ws/src/f1tenth_gym_ros/maps/levine_closed_static_obstacle'

Edit the obstacle_type variable value to static_obstacles, while keeping others fixed. Build the workspace and do the same steps as in explained in Part 2.

4) Residual RL (SAC) Controller - Dynamic Obstacles - levine_closed map

Do the following changes in sim.yaml

# map parameters
map_name: 'levine_closed'
map_path: '/sim_ws/src/f1tenth_gym_ros/maps/levine_closed'

# opponent parameters
num_agent: 2

Do the following changes in RL/test.py

env = F110Gym(is_train=False, is_opp=True)
obs = env.reset()

exp_no = 1
algo = "SAC"
logs_type = "obs_logs"
obstacle_type="dynamic_obstacles_2"

Ensure that the is_opp variable is set to True. Build the workspace and do the same steps as in explained in Part 2. Now you can see two racecars spawn. One runs the Simple Pure Pursuit and the other runs the Residual RL controller.