Beyond Safe Execution: Safe Online Reinforcement Learning for Autonomous Driving with Value Calibration and Adaptive Guidance

The paper presents SOAR-Flow, a framework for safe and stable online fine-tuning of offline-initialized Flow-based driving policies. It integrates a control barrier function with model predictive control to filter unsafe actions, and uses value calibration and adaptive guidance to improve policy performance while reducing the need for safety corrections. The approach is evaluated in simulation and on an uncrewed ground vehicle (UGV), showing improved driving performance and reduced reliance on safety corrections during online fine-tuning.