Enhancing Autonomous Vehicle Training with Language Model Integration and Critical Scenario Generation
This work addresses the problem of improving robustness and safety in autonomous vehicle systems for developers and researchers, though it appears incremental as it builds on existing RL methods with added scenario generation and optional LLM integration.
The paper tackles the challenge of training autonomous vehicles by introducing CRITICAL, a closed-loop framework that generates diverse critical driving scenarios targeting specific learning gaps in RL agents, resulting in noticeable performance improvements in evaluations using PPO and HighwayEnv.
This paper introduces CRITICAL, a novel closed-loop framework for autonomous vehicle (AV) training and testing. CRITICAL stands out for its ability to generate diverse scenarios, focusing on critical driving situations that target specific learning and performance gaps identified in the Reinforcement Learning (RL) agent. The framework achieves this by integrating real-world traffic dynamics, driving behavior analysis, surrogate safety measures, and an optional Large Language Model (LLM) component. It is proven that the establishment of a closed feedback loop between the data generation pipeline and the training process can enhance the learning rate during training, elevate overall system performance, and augment safety resilience. Our evaluations, conducted using the Proximal Policy Optimization (PPO) and the HighwayEnv simulation environment, demonstrate noticeable performance improvements with the integration of critical case generation and LLM analysis, indicating CRITICAL's potential to improve the robustness of AV systems and streamline the generation of critical scenarios. This ultimately serves to hasten the development of AV agents, expand the general scope of RL training, and ameliorate validation efforts for AV safety.