Trajectory Tracking Control of Unmanned Surface Vessel Based on Neural Network Observer

In this paper, we propose a trajectory tracking control scheme for unmanned ships based on neural network observers, which has model uncertainty, unknown environmental disturbance and saturation problems. A neural network-based observer was developed to reconstruct unmeasured velocity and estimate the uncertainty of the model. Using the neural network, a neural adaptive output feedback controller was developed. In addition, a stable controller was designed by the backstepping method. Finally, the Lyapunov analysis shows that all signals in the closed-loop system are bounded. The feasibility of the proposed control scheme is verified by simulation.