Luke Rowe*, Roger Girgis*, Anthony Gosselin, Bruno Carrez, Florian Golemo, Felix Heide, Liam Paull, Christopher Pal
We propose CtRL-Sim, a framework that leverages offline reinforcement learning for controllable closed-loop multi-agent driving simulation.
Code will be released soon.
Check LICENSE
@InProceedings{
rowe2024ctrlsim,
title={{CtRL-Sim}: Reactive and Controllable Driving Agents with Offline Reinforcement Learning},
author={Luke Rowe and Roger Girgis and Anthony Gosselin and Bruno Carrez and Florian Golemo and Felix Heide and Liam Paull and Christopher Pal},
journal={arXiv},
volume={2403.19918},
year={2024}
}