Event-Triggered Fault Detection Filter Design for Autonomous Underwater Vehicles

The problem of event-triggered fault detection (FD) for autonomous underwater vehicles (AUVs) subject to actuator faults is addressed in this paper. Firstly, applying a linear discrete-time system with varying sampling periods, one can obtain the dynamics of an event-triggered system. Then, using an observer-based fault detection filter (FDF) as the residual generator, the generated residual signal and the event-triggered transmission errors can be decoupled completely. Next, a multiobjective formulation of the event-triggered FD problem is given in the context of H
i
/$H_{\infty}$ optimization. An optimal solution to the event-triggered H
i
/$H_{\infty}$-FDF design issue is then obtained by calculating Riccati equations recursively. It is shown that the proposed event-triggered FD scheme can achieve satisfactory the FD performance, while the amount of data transmitted to the FD module is reduced significantly. Simulation results based on an AUV demonstrate the effectiveness of the proposed method.