A game-theoretic approach of cyberattack resilient constraint-following control for cyber-physical systems

Cyber-physical systems (CPSs) integrate network and physical components. The relevant network security issues have been receiving consistent attention. This paper considers cyber-physical systems whose control resilience is characterized by uncertainty, mechanical motion constraints, and exposure to cyber-attacks. To address these problems, this paper proposes a game-theoretic method based on constraint following theory where the moving trajectory requirements are considered as constraints. There can be adversary network environment which may launch cyberattacks and even deny providing service. With a creative control design, it is validated that the controlled system is resilient to cyberattack disturbance and other spoofing or mixed threat attacks. Three game rules, one non-cooperative game and two Stackelberg competitions, are adopted in this study to obtain the optimal control design parameter choice. It is proven that the optimal choice is the same from the three rules, thus making this optimal choice generic (i.e., nonspecific) and important. The superiority of the optimal control design parameter is demonstrated in simulations.