Robust control strategy for platoon of connected and autonomous vehicles considering falsified information injected through communication links

Abstract

Connected and Autonomous Vehicles (CAVs) in a platoon can exchange real-time information using Vehicle-to-Vehicle (V2V) communication technology to enhance platoon control performance. However, the V2V communication technology also provides opportunities for cyber-attacks, where falsified information can be injected into vehicle controllers to disrupt the platoon operation and even induce vehicle collisions. To address this problem, this study proposes a robust platoon control strategy for CAVs to mitigate the impacts of the falsified information to maneuver the CAV platoon to achieve consensus safely. The proposed control strategy consists of three components: (i) a $H_{\infty }$ robust control law, which consistently negates the disturbance induced by falsified information; (ii) a state observer which estimates the vehicle states and disturbance induced by falsified information and inputs the estimated results into the $H_{\infty }$ robust control law to compute a synthesized control decision; and (iii) a control decision regulator which applies a Control Barrier Function-based Quadratic Programming (CBF-QP) to regulate the control decision computed by the $H_{\infty }$ robust control law to avoid actuator saturation issue and ensure safe spacing for each vehicle in the platoon. Numerical experiments demonstrate that the proposed control strategy can effectively drive the CAV platoon to the desired consensus safely and efficiently under the impacts of falsified information injection.