Asynchronous Fault Detection for Unmanned Marine Vehicles under Aperiodic DoS Attacks and Stochastic Deception Attacks
TL;DR: This paper focuses on asynchronous thruster fault detection for unmanned marine vehicles (UMVs) in the presence of multiple cyber threats, external disturbances, and thruster failures.
Abstract: This paper focuses on asynchronous thruster fault detection for unmanned marine vehicles (UMVs) in the presence of multiple cyber threats, external disturbances, and thruster failures. A novel detection model is developed utilizing an asynchronous switched method, specifically addressing aperiodic Denial-of-Service (DoS) attacks and stochastic Deception attacks. The proposed approach employs the Lyapunov–Krasovskii functional method combined with model-dependent average dwell time to derive a set of sufficient conditions that ensure the system achieves global mean-square exponential stability with a guaranteed ${H_\infty }$ performance. Furthermore, the research determines the tolerable upper and lower bounds for constrained DoS attacks, enhancing the system's robustness against such cyber threats. Solvable conditions for the design of fault detection filters are derived using decoupling techniques, ensuring the efficacy of the detection mechanism. Finally, extensive simulations on a UMV validate the proposed method's feasibility and effectiveness, demonstrating its capability to maintain reliable fault detection under complex and evolving cyber attack scenarios.
Submission Number: 250
Loading