Expert Knowledge-driven Reinforcement Learning for Autonomous Racing via Trajectory Guidance and Dynamics Constraints
This research provides an incremental improvement for autonomous racing systems by enhancing the safety and performance of reinforcement learning agents.
This paper addresses the instability and unsafe actions of reinforcement learning in autonomous racing by proposing TraD-RL, a method that integrates expert racing line knowledge and vehicle dynamics constraints. The method improves both lap speed and driving stability, achieving a synergistic optimization of racing performance and safety in a high-fidelity simulation.
Reinforcement learning has demonstrated significant potential in the field of autonomous driving. However, it suffers from defects such as training instability and unsafe action outputs when faced with autonomous racing environments characterized by high dynamics and strong nonlinearities. To this end, this paper proposes a trajectory guidance and dynamics constraints Reinforcement Learning (TraD-RL) method for autonomous racing. The key features of this method are as follows: 1) leveraging the prior expert racing line to construct an augmented state representation and facilitate reward shaping, thereby integrating domain knowledge to stabilize early-stage policy learning; 2) embedding explicit vehicle dynamic priors into a safe operating envelope formulated via control barrier functions to enable safety-constrained learning; and 3) adopting a multi-stage curriculum learning strategy that shifts from expert-guided learning to autonomous exploration, allowing the learned policy to surpass expert-level performance. The proposed method is evaluated in a high-fidelity simulation environment modeled after the Tempelhof Airport Street Circuit. Experimental results demonstrate that TraD-RL effectively improves both lap speed and driving stability of the autonomous racing vehicle, achieving a synergistic optimization of racing performance and safety.