Cyber-Physical-Level Safe Control of Intelligent and Connected Vehicles With Unreliable Communication and Actuator Faults

The utilization of vehicle-to-vehicle (V2V) communication techniques within a vehicle platooning system (VPS) significantly enhances traffic safety and efficiency by exchanging information. However, packet loss can cause V2V communication to be unreliable, leading to the switching of communication topology and threatening the safety of the VPS. Furthermore, in practical scenarios, intelligent and connected vehicles (ICVs) inevitably suffer inherent unknown actuator faults that impose potential safety risks on a VPS. In this work, we propose a novel cyber-physical-level safe control framework that consists of an upper-level cyber plane and a lower-level physical plane. Within the cyber plane, we use a continuous-time Markov chain to model switching communication topology and design a fully distributed adaptive observer for each following vehicle to obtain the leader's state. Based on adaptive fault-tolerant controllers in the physical plane, each ICV can track its observer state and compensate for the influences of unknown actuator faults. Moreover, several sufficient conditions are derived to guarantee the mean square stability of the VPS and achieve the desired control objectives. Finally, simulation results are provided to verify the effectiveness of our proposed control method and to demonstrate its competitiveness.