Finite-Time Based Fault-Tolerant Control for Half-Car Active Suspension System With Cyber-Attacks: A Memory Event-Triggered Approach

By wielding memory event-triggered scheme (METS), this paper explores the issue of finite-time fault-tolerant control strategy for a sort of half-car active suspension system, which comprises actuator failures and cyber-attacks. Primarily, a full-order state observer is modeled to estimate the immeasurable states of the system. Based upon this observer design, a fault-tolerant controller is well-established to confront the partial loss and to adopt the actuator fault inevitably. Subsequently, a proper METS is proffered to mitigate the frequent packet transmission in the network channel. In contrary with the conventional event-triggered scheme (ETS), the prescribed METS can produce some newest triggering events by utilizing the current released packets. Thereby, the event generator can make more precise determinations, resulting in an improved control performance. Notably, the cyber-attack phenomenon is described by the Bernoulli distributed stochastic variable. By engaging Lyapunov function and integral inequality technique, sufficient condition in the structure of linear matrix inequality (LMI) assure the asymptotic mean-square finite-time boundedness (AMFTB) of the half-car active suspension model with a prescribed $H_\infty$ performance attenuation. Moreover, the controller and observer gain matrices are stipulated from the attained LMI conditions. Ultimately, the capability of the proposed control design is demonstrated through a half-car active suspension model.