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

TL;DR

TorchDriveEnv is a flexible, Python-based reinforcement learning benchmark for autonomous driving that features reactive, realistic, and diverse NPCs, enabling comprehensive testing of vehicle behavior under various conditions.

Contribution

It introduces TorchDriveEnv, a customizable, integrated simulation environment for autonomous driving research with reactive NPCs, enhancing realism and ease of use.

Findings

Efficient and simple to use for training and evaluation

Reveals the difficulty of solving complex driving tasks

Supports testing of various kinematic models and traffic patterns

Abstract

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 behavior, 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 behavioral simulation API. This allows users to train and evaluate driving models alongside data driven Non-Playable Characters (NPC) whose initializations and driving behavior 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.