Abstract: This study aims to solve the problem of trajectory tracking control for underactuated surface vessels (USVs) suffering from unmodeled dynamics and external disturbances. Firstly, the tracking errors system is divided into two channels for controller design through the model transformation. Then, in order to accommodate unknow system dynamics and external perturbation, the Minimum-Learning-Parameter based neural network are adopted to compensate the uncertainties. After that, a fixed-time sliding mode controller is proposed with the application of hyperbolic tangent function, which can ensure fast convergence of tracking errors. Numerical simulation results illustrate that the fixed-time convergence for tracking errors can be guaranteed independent of initial states. The designed control scheme not only possesses fixed-time convergence rate, but also shows satisfactory disturbance rejection capability.