A Simulation Benchmark for Autonomous Racing with Large-Scale Human Data
This work addresses the need for realistic and accessible testing environments for autonomous racing algorithms, benefiting researchers in robotics and AI, though it is incremental as it builds on existing simulators and methods.
The authors tackled the problem of limited research in autonomous racing due to high costs and inaccurate simulators by proposing a racing simulation platform based on Assetto Corsa, which includes state-of-the-art algorithms and a human dataset, resulting in publicly released code and benchmarks.
Despite the availability of international prize-money competitions, scaled vehicles, and simulation environments, research on autonomous racing and the control of sports cars operating close to the limit of handling has been limited by the high costs of vehicle acquisition and management, as well as the limited physics accuracy of open-source simulators. In this paper, we propose a racing simulation platform based on the simulator Assetto Corsa to test, validate, and benchmark autonomous driving algorithms, including reinforcement learning (RL) and classical Model Predictive Control (MPC), in realistic and challenging scenarios. Our contributions include the development of this simulation platform, several state-of-the-art algorithms tailored to the racing environment, and a comprehensive dataset collected from human drivers. Additionally, we evaluate algorithms in the offline RL setting. All the necessary code (including environment and benchmarks), working examples, datasets, and videos are publicly released and can be found at: https://assetto-corsa-gym.github.io