Attack-Resilient Lateral Stability Control for Autonomous In-Wheel-Motor-Driven Electric Vehicles

Abstract

This paper presents an attack-resilient lateral stability control design approach for autonomous in-wheel-motor-driven electric vehicles (EVs) to enhance the cyber-physical security of the steering system. Specifically, this work examines cyber-attacks on the steering angle, one of the most safety-critical control signals in autonomous driving systems. Different from the current literature, the proposed system serves as an assistant controller in the monitoring system. Once a cyber-attack is identified, a disturbance estimator is then designed based on the lateral system dynamics, which models cyber-attacks as an exogenous disturbance. Then, a robust predictive controller is developed to mitigate the impact of cyber-attacks and ensure driving safety. The proposed attack-resilient control carries a low computational burden, making it applicable for real-time applications. Simulation results are finally provided to demonstrate the effectiveness of the proposed method.