Event-Triggered Finite-Time Tracking Control of Unmanned Surface Vessel Using Neural Network Prescribed Performance

The current paper verifies the event-triggered finite-time tracking control of a fully actuated unmanned surface vessel under unmodeled dynamics and external disturbances. Radial basis function neural networks (RBFNNs), nonlinear disturbance observers (NDO), and the event-triggered mechanism (ETM) are utilized to design a new type of finite time tracking controller (FFTC). The proposed controller utilizes the dynamic surface control (DSC) approach to resolve the “differential explosion” issue of virtual control laws. Prescribed performance functions (PPFs) and the finite-time control (FFC) technique are utilized to specify the efficiency of tracking errors. The designed control laws make the control system of USV semi-globally practically finite-time stable (SGPFS) and make the tracking errors tend to a narrow residual set involving the specified bound in a finite time. In the end, the simulations reflect the presented FFTC’s efficiency.