Resilient Control of Networked Control Systems With Hidden DoS Attacks and Unknown Observation Probability

Abstract

This article systematically studies the problem of resilient controller design for networked control systems under imperfect communication environments. The stochastic denial of service attacks in the case of an imperfect network are remodeled. An semi-Markov chain is introduced in this article to provide a suitable description of random attacks. On this basis, the double-layer stochastic process is constructed, which consists of an attack sequence and an observation one. Second, since the attack modes are inaccessible, a controller is constructed based on the observation sequence by virtue of the emission probability to stabilize the system under random attacks. Due to the complexity of the network, the probability that an observer can accurately detect an attack when it occurs is considered unknown. Based on the introduced double-layer stochastic process, a mode-associated Lyapunov function is constructed to facilitate the stability analysis. Building on this foundation, sufficient conditions that can ensure the $\delta$-error mean-square stability of the closed-loop system are derived. In the end, a simulation of the autonomous ground vehicle is presented, showcasing the feasibility and validity of the proposed control scheme.