RLfOLD

This is the implementation of RLfOLD, which is described in:

RLfOLD: Reinforcement Learning from Online Demonstrations in Urban Autonomous Driving

Daniel Coelho, Miguel Oliveira, Vítor Santos.

AAAI 2024

If you find our work useful, please consider citing:

@inproceedings{coelho2024rlfold,
  title={RLfOLD: Reinforcement Learning from Online Demonstrations in Urban Autonomous Driving},
  author={Coelho, Daniel and Oliveira, Miguel and Santos, Vitor},
  booktitle={Proceedings of the AAAI Conference on Artificial Intelligence},
  volume={38},
  number={10},
  pages={11660--11668},
  year={2024}
}

Setup

• Clone the repository with git clone git@github.com:DanielCoelho112/rlfold.git
• Download CARLA 0.9.10.1.
• Run the docker container with docker run -it --gpus all --network=host -v results_path:/root/results/rlfold -v rlfold_path:/root/rlfold danielc11/rlfold:0.0 bash where results_path is the path where the results will be written, and rlfold_path is the path of the rlfold repository.

Training

• Start the CARLA server
• Run: python3 rlfold/run/python3 main.py -en rlfold_original

Credits

Thanks to the authors of End-to-End Urban Driving by Imitating a Reinforcement Learning Coach for providing a framework to train RL agent in CARLA.