Secure platooning control of automated vehicles under cyber attacks

• Published in: ISA transactions Vol. 127; pp. 229 - 238
• Main Authors: Mousavinejad, Eman, Vlacic, Ljubo
• Format: Journal Article
• Language: English
• Published: United States Elsevier Ltd 01.08.2022
• Subjects: Cyber attacks; Secure control; Set-membership estimation; Vehicle platooning; Vehicular ad-hoc networks (VANETs); Cyber attacks; Set-membership estimation; Vehicle platooning; Vehicular ad-hoc networks (VANETs); Secure control
• ISSN: 0019-0578; 1879-2022; 1879-2022
• DOI: 10.1016/j.isatra.2022.02.046

This paper studies the problem of secure control of automated vehicles in a platoon-based driving pattern over a vehicular ad-hoc network (VANET) subject to various cyber attacks. The platoon under consideration is a convoy of a leader vehicle whose control input is unknown to its following vehicles and some follower vehicles with uncertain heterogeneous engine time constants, bounded disturbance and noise. First, a local estimator is developed for each follower vehicle so as to construct some confidence ellipsoidal estimation region always enclosing vehicular true state regardless of uncertain heterogenous engine time constants, bounded disturbance and noise. A convex optimization algorithm is proposed to find some optimal ellipsoidal sets and recursively solve out gain matrices of the local estimators. Then, a scalable control protocol employing the state estimates from its local and underlying neighboring estimators is designed to accomplish secure platooning control. Under the derived design technique, the resulting closed-loop platooning tracking errors are proven to remain in the vicinity of zero. Comparative studies are conducted to validate the efficacy of the proposed control method on achieving the satisfactory platooning performance by handling different attack strategies. •A scalable secure platooning control protocol is designed for automated vehicles.•A class of amplitude-bounded attacks is considered to represent various attacks.•A local set-membership estimator is designed for each follower vehicle.•A general model of vehicle longitudinal dynamics is considered.