Hierarchical Control for Cooperative Teams in Competitive Autonomous Racing

TL;DR

This paper presents a hierarchical control framework for cooperative autonomous racing that effectively manages complex racing rules and demonstrates superior performance and human-like strategic behavior compared to baseline methods.

Contribution

It introduces a generalized hierarchical control approach with a high-level tactical planner and two low-level path planning methods, including MARL and LQNG, for competitive autonomous racing.

Findings

Hierarchical controllers outperform baselines in race wins and team performance.

Controllers mimic expert human driving behaviors like overtaking and defending.

The approach effectively handles complex racing rules and strategic planning.

Abstract

We investigate the problem of autonomous racing among teams of cooperative agents that are subject to realistic racing rules. Our work extends previous research on hierarchical control in head-to-head autonomous racing by considering a generalized version of the problem while maintaining the two-level hierarchical control structure. A high-level tactical planner constructs a discrete game that encodes the complex rules using simplified dynamics to produce a sequence of target waypoints. The low-level path planner uses these waypoints as a reference trajectory and computes high-resolution control inputs by solving a simplified formulation of a racing game with a simplified representation of the realistic racing rules. We explore two approaches for the low-level path planner: training a multi-agent reinforcement learning (MARL) policy and solving a linear-quadratic Nash game (LQNG) approximation. We evaluate our controllers on simple and complex tracks against three baselines: an end-to-end MARL controller, a MARL controller tracking a fixed racing line, and an LQNG controller tracking a fixed racing line. Quantitative results show our hierarchical methods outperform the baselines in terms of race wins, overall team performance, and compliance with the rules. Qualitatively, we observe the hierarchical controllers mimic actions performed by expert human drivers such as coordinated overtaking, defending against multiple opponents, and long-term planning for delayed advantages.