Cyber-physical systems under hybrid cyber-attacks: Resilient event-triggered H∞ control approach.

Abstract

This article introduces a novel adaptive event-triggered finite-time H_∞ control approach for Cyber-Physical Systems (CPSs) under the influence of bounded disturbances, actuator faults, and randomly occurring hybrid cyber-attacks, including aperiodic DoS attacks, deception attacks, and replay attacks. This paper presents a novel integrated framework for hybrid attacks that consists of aperiodic DoS attacks, deception attacks, and replay attacks. To minimize continuous data transmissions in the network and maximize the utilization of limited resources, an adaptive periodic event-triggered mechanism (APETM) is developed by employing the DoS-dependent adaptive varying threshold. Compared to common event-triggered mechanisms, this method allows for periodic monitoring and effectively increases the lower bound of event intervals, making it well-suited for real-world applications. Initially, a new theoretical framework for cyber-physical systems is proposed, incorporating an adaptive periodic event-triggered mechanism under hybrid cyber-attacks, external disturbances, and actuator faults. Based on the established model, the Lyapunov-Krasovskii stability theory is employed to establish sufficient conditions for ensuring the finite-time stability with an H_∞ performance level. Additionally, a linear matrix inequality technique is utilized to determine the controller gains and triggering parameters. The effectiveness of the proposed method is demonstrated through a simulation of the one-link flexible joint robot systems.