Securing End-to-End Reinforcement Learning-Driven Autonomous Driving: A Control Command Utility-based Intrusion Response System

End-to-end autonomous driving with deep reinforcement learning (DRL) encounters significant challenges in security and safety, which are critical for the automotive industry's adoption of autonomous technologies. This paper introduces a novel intrusion response system (IRS), called the control command utility-based IRS (CCU), specifically designed for DRLbased autonomous systems. The CCU provides a lightweight yet powerful defense against false data injection attacks on the invehicle CAN (control area network) bus, enhancing both security and driving performance by making intelligent, context-aware decisions based on control command utilities derived from DRL outputs. We rigorously evaluated CCU against other state-ofthe-art IRSs based on DRL autonomous driving models, Rails and Roach. Equipped with an additional confidence score-based filter, CCU effectively minimizes false alarms, demonstrating superior performance in improving critical driving metrics such as driving score, route completion, and infraction penalties, all while lowering defense costs. Furthermore, CCU exhibits robust resilience in hostile environments with varying attack probabilities, underscoring its reliability in complex scenarios. This contribution represents a significant advancement in autonomous driving, addressing essential security and safety challenges and accelerating the path toward safer, more reliable autonomous vehicle deployment.