In this paper, output feedback control is investigated for a class of ship course discrete- time uncertain nonlinear system with rudder dynamics in strict-feedback form. Firstly, the original states strict-feedback system is transformed into a cascade form. Then with employment of the control input and course output, the adaptive neural network controller is developed for the output feedback system. The proposed algorithm is simple, it is feasible when some system states are unmeasurable, herein, it is convenient to implemented in application. In addition, It is proven that all the signals in closed-loop system are semi-globally uniformly ultimately bounded (SGUUB), and the tracking error converges to a small neigborhood of the origin. At the last, a simulation of motor vessel “yukun” is employed to illustrate the effectiveness of the proposed algorithem.