Head-to-Head autonomous racing at the limits of handling in the A2RL challenge

TL;DR

This paper details the algorithms and strategies used by TUM Autonomous Motorsport to excel in autonomous racing, emphasizing handling limits, multi-vehicle interactions, and human-like driving behavior in the A2RL challenge.

Contribution

It introduces novel algorithms and deployment strategies for autonomous racing that emulate human driving and operate at the vehicle handling limits in a competitive environment.

Findings

Successfully won the A2RL with advanced handling and interaction strategies.

Demonstrated the effectiveness of human-like driving behavior in autonomous racing.

Identified key factors enabling high-performance autonomous vehicle control.

Abstract

Autonomous racing presents a complex challenge involving multi-agent interactions between vehicles operating at the limit of performance and dynamics. As such, it provides a valuable research and testing environment for advancing autonomous driving technology and improving road safety. This article presents the algorithms and deployment strategies developed by the TUM Autonomous Motorsport team for the inaugural Abu Dhabi Autonomous Racing League (A2RL). We showcase how our software emulates human driving behavior, pushing the limits of vehicle handling and multi-vehicle interactions to win the A2RL. Finally, we highlight the key enablers of our success and share our most significant learnings.