Resilient and robust control strategy against deception attacks in the platoon of vehicles

• Published in: IET control theory & applications 2024-12, Vol.18 (18), p.2830-2840
• Main Authors: Mokari, Hassan, Firouzmand, Elnaz, Sharifi, Iman, Doustmohammadi, Ali
• Format: Article
• Language: English
• Subjects: distributed control; multi‐agent systems; robust control; vehicles
• ISSN: 1751-8644; 1751-8652
• DOI: 10.1049/cth2.12757

A platoon must ensure the security of conducting missions due to the susceptibility of the vehicles to cyber‐attacks. The deception attack is a common type of cyber‐attack that might be introduced during data exchange through communication channels. Therefore, it can divert each vehicle by injecting uncertainty into the communication link between the agents in the cyber domain. This paper employs a leader–follower consensus system as a reference model to control the platoon vehicles. Accordingly, each agent in the reference model sends its information as a reference trajectory to its corresponding vehicle within the platoon. The vehicles are equipped with a robust local controller, enabling them to follow their desired trajectory in the presence of external disturbances. Furthermore, each vehicle employs an unknown input observer to detect a deception attack while decoupling it from external disturbances. Moreover, by introducing a switching strategy based on mode‐dependent average dwell time to the leader–follower consensus system, a novel resilient control strategy against the deception attack is implemented in the cyber domain. The resultant resilient leader–follower consensus system in the platoon retrieves and restores the attacked vehicle in the platoon to its associated states. The applicability of the proposed method is shown via simulation. 1)A leader–follower consensus system is employed as a reference model for controlling the platoon vehicles by sharing the reference trajectories. 2)Each vehicle is equipped with a robust local controller to maintain the given reference path despite disturbances. 3)Vehicles employ unknown input observers to detect deception attacks and isolate them from external disturbances. After the detection phase and through a switching strategy based on mode‐dependent average dwell time, resilience against deception attacks in the cyber domain is achieved, ensuring the recovery of attacked vehicles to their intended states within the platoon.