High-speed Autonomous Racing using Trajectory-aided Deep Reinforcement Learning

2023/06/12 by Benjamin Evans, Evans, Benjamin David, Herman A. Engelbrecht +3 · 6 citations
Computer Science · Economics, Econometrics and Finance · #FOS: Computer and information sciences #Reinforcement Learning in Robotics #Robotics (cs.RO) #Software Testing and Debugging Techniques #Sports Analytics and Performance

paper · pdf · doi:10.48550/arxiv.2306.07003

openalex publication_date 2023/06/12 · openalex created_date 2023/06/14 · openalex updated_date 2026/07/28

Abstract

The classical method of autonomous racing uses real-time localisation to follow a precalculated optimal trajectory. In contrast, end-to-end deep reinforcement learning (DRL) can train agents to race using only raw LiDAR scans. While classical methods prioritise optimization for high-performance racing, DRL approaches have focused on low-performance contexts with little consideration of the speed profile. This work addresses the problem of using end-to-end DRL agents for high-speed autonomous racing. We present trajectory-aided learning (TAL) that trains DRL agents for high-performance racing by incorporating the optimal trajectory (racing line) into the learning formulation. Our method is evaluated using the TD3 algorithm on four maps in the open-source F1Tenth simulator. The results demonstrate that our method achieves a significantly higher lap completion rate at high speeds compared to the baseline. This is due to TAL training the agent to select a feasible speed profile of slowing down in the corners and roughly tracking the optimal trajectory.

Cited by

• Context-Aware Model-Based Reinforcement Learning for Autonomous Racing
• End2Race: Efficient End-to-End Imitation Learning for Real-Time F1Tenth Racing
• IteraOptiRacing: A Unified Planning-Control Framework for Real-time Autonomous Racing for Iterative Optimal Performance
• MIND-Stack: Modular, Interpretable, End-to-End Differentiability for Autonomous Navigation
• Expert Knowledge-driven Reinforcement Learning for Autonomous Racing via Trajectory Guidance and Dynamics Constraints
• Drive Fast, Learn Faster: On-Board RL for High Performance Autonomous Racing

Related

• End-to-end Lidar-Driven Reinforcement Learning for Autonomous Racing
• Drive Fast, Learn Faster: On-Board RL for High Performance Autonomous Racing
• End2Race: Efficient End-to-End Imitation Learning for Real-Time F1Tenth Racing
• Sense, Imagine, Act: Multimodal Perception Improves Model-Based Reinforcement Learning for Head-to-Head Autonomous Racing
• Autonomous Racing using a Hybrid Imitation-Reinforcement Learning Architecture