DriverGym: Democratising Reinforcement Learning for Autonomous Driving

2021/11/12 by Parth Kothari, Kothari, Parth, Christian Perone +9 · 2 citations
Computer Science · Engineering · #Artificial intelligence #Autonomous Vehicle Technology and Safety #Cloning (programming) #Code (set theory) #Computer Vision and Pattern Recognition (cs.CV) #Computer science #FOS: Computer and information sciences #Machine Learning (cs.LG) #Machine learning #Open source #Operating system #Protocol (science) #Reinforcement Learning in Robotics #Reinforcement learning #Software #Traffic control and management #cs.CV #cs.LG

paper · pdf · doi:10.48550/arxiv.2111.06889

published in arXiv (Cornell University) (Cornell University) · Accepted to NeurIPS 2021 ML4AD Workshop

arxiv created 2021/11/12 · openalex publication_date 2021/11/12 · arxiv updated 2021/11/16 · openalex created_date 2025/10/10 · openalex updated_date 2026/08/04

Abstract

Despite promising progress in reinforcement learning (RL), developing algorithms for autonomous driving (AD) remains challenging: one of the critical issues being the absence of an open-source platform capable of training and effectively validating the RL policies on real-world data. We propose DriverGym, an open-source OpenAI Gym-compatible environment specifically tailored for developing RL algorithms for autonomous driving. DriverGym provides access to more than 1000 hours of expert logged data and also supports reactive and data-driven agent behavior. The performance of an RL policy can be easily validated on real-world data using our extensive and flexible closed-loop evaluation protocol. In this work, we also provide behavior cloning baselines using supervised learning and RL, trained in DriverGym. We make DriverGym code, as well as all the baselines publicly available to further stimulate development from the community.

Citations

• Mastering Chess and Shogi by Self-Play with a General Reinforcement Learning Algorithm
• A Reduction of Imitation Learning and Structured Prediction to No-Regret Online Learning
• ChauffeurNet: Learning to Drive by Imitating the Best and Synthesizing the Worst
• SMARTS: Scalable Multi-Agent Reinforcement Learning Training School for Autonomous Driving

Cited by

• UrbanVerse: Scaling Urban Simulation by Watching City-Tour Videos
• TreeIRL: Safe Urban Driving with Tree Search and Inverse Reinforcement Learning
• Towards Autonomous Micromobility through Scalable Urban Simulation

Related

• Exploring applications of deep reinforcement learning for real-world autonomous driving systems
• An Independent Study of Reinforcement Learning and Autonomous Driving
• V-Max: A Reinforcement Learning Framework for Autonomous Driving
• RAD: Training an End-to-End Driving Policy via Large-Scale 3DGS-based Reinforcement Learning
• Deep Reinforcement Learning for Autonomous Driving: A Survey