Aiming at the problems of dynamic uncertainty and unknown disturbance in the trajectory tracking control of underactuated surface ships, an adaptive neural network finite time trajectory tracking control scheme is designed.

The underactuated variation is carried out using the method of kinematic virtual control law transformation and bounded constraints. Under the framework of Backstepping, neural networks are used to reconstruct unknown dynamics, and an adaptive law is designed to approach the upper bound of unknown disturbances. The Lyapunov direct method provides a rigorous theoretical analysis which proves that all the signals of the closed-loop system are bounded, and the tracking error converges to a bounded interval.

The simulation results show that this control scheme can make an underactuated ship track the desired trajectory in a limited time, the convergence speed of the system error is faster than that of the traditional control scheme, and the upper and lower bounds of the error are also smaller. It also shows good robustness in the face of unknown time-varying interference from the outside world.

The results of this study can provide valuable references for the tracking and control of ship trajectories, giving it great practical engineering significance.