J. Wilder Lavington

J. Wilder Lavington

Applied Scientist working on automated negotiation, reinforcement learning, and large language models.

TorchDriveEnv: A Reinforcement Learning Benchmark for Autonomous Driving with Reactive, Realistic, and Diverse Non-Playable Characters

J. Wilder Lavington, Ke Zhang, Vasileios Lioutas, Matthew Niedoba, Yunpeng Liu, Dylan Green, Saeid Naderiparizi, Xiaoxuan Liang, Setareh Dabiri, Adam Scibior, Berend Zwartsenberg, Frank Wood

arXiv 2024 arXiv preprint ·

Figure 1 from the paper
Figure 1 from the paper.

The training, testing, and deployment, of autonomous vehicles requires realistic and efficient simulators. Moreover, because of the high variability between different problems presented in different autonomous systems, these simulators need to be easy to use, and easy to modify. To address these problems we introduce TorchDriveSim and its benchmark extension TorchDriveEnv. TorchDriveEnv is a lightweight reinforcement learning benchmark programmed entirely in Python, which can be modified to test a number of different factors in learned vehicle behaviour, including the effect of varying kinematic models, agent types, and traffic control patterns. Most importantly unlike many replay based simulation approaches, TorchDriveEnv is fully integrated with a state of the art behavioural simulation API. This allows users to train and evaluate driving models alongside data driven Non-Playable Characters (NPC) whose initializations and driving behaviour are reactive, realistic, and diverse. We illustrate the efficiency and simplicity of TorchDriveEnv by evaluating common reinforcement learning baselines in both training and validation environments. Our experiments show that TorchDriveEnv is easy to use, but difficult to solve.

BibTeX

@misc{lavington2024torchdriveenvreinforcementlearningbenchmark,
      title={TorchDriveEnv: A Reinforcement Learning Benchmark for Autonomous Driving with Reactive, Realistic, and Diverse Non-Playable Characters},
      author={Jonathan Wilder Lavington and Ke Zhang and Vasileios Lioutas and Matthew Niedoba and Yunpeng Liu and Dylan Green and Saeid Naderiparizi and Xiaoxuan Liang and Setareh Dabiri and Adam Ścibior and Berend Zwartsenberg and Frank Wood},
      year={2024},
      eprint={2405.04491},
      archivePrefix={arXiv},
      primaryClass={cs.AI},
      url={https://arxiv.org/abs/2405.04491}