Robust Decentralized Control for Local Detection of Covert Cyberattacks in Interconnected Systems

Interconnected systems consist of multiple subsystems coupled through physical and cyber connections. Within a subsystem, a covert cyberattack manipulates actuator commands to achieve a malicious objective while simultaneously manipulating sensor measurements to mimic nominal behavior, thereby remain undetected. Traditionally in the literature, approaches are proposed to detect and isolate such cyberattacks by using banks of observers, irrespective of any control scheme. In addition, the detection of a covert cyberattack within a subsystem as well as its isolation is feasible only by its neighbouring subsystems. In this paper, we propose an alternative approach, by incorporating a decentralized controller and a decentralized state-disturbance observer within each subsystem, which makes detection of covert cyberattacks feasible locally within each subsystem (i.e., local detection), thereby eliminating the reliance on neighbouring subsystems as well as the need for additional design complexity for isolation. The proposed approach is validated through numerical simulations.