Composite neural learning event-triggered control for dynamic positioning vehicles with the fault compensation mechanism

This paper focus on the novel composite neural learning event-triggered control for dynamic positioning vehicles with the fault compensation mechanism. In the proposed algorithm, the system uncertainties are tackled with by incorporating the dynamic surface control (DSC) and the robust neural damping techniques. Besides, the serial-parallel estimation model (SPEM) is developed to produce the prediction errors, which could be employed to derive the corresponding composite adaptive law. And the unknown actuator faults and gain uncertainties are effectively compensated for merits of the composite intelligent learning method. Furthermore, the idea of relative threshold strategy is utilized to construct the event-triggered mechanism. The event-triggered input could reduce the communication burden in the channel from controller to actuators. Through the direct Lyapunov theory, the parameters setting can be derived to guarantee the semi-global uniformly ultimately bounded (SGUUB) stability of all error signals in the closed-loop system. Finally, the effectiveness of the proposed algorithm is validated through the simulation experiments.