The dynamic positioning system is a system that ships rely on their own propulsion devices to maintain position and course. Pod propulsion was more and more installed in dynamic positioning unmanned surface vessels because it can rotate 360 degrees freely. The maximum disturbance of unmanned surface vessel dynamic positioning comes from the forces and moments generated by the ocean wind, wave and current. In order to improve its accuracy of disturbance rejection, the mathematical model of the vessel was established according to the special structure of pod propulsion ship, and the calculations of the forces and moments of wind, wave and current are given. Then, the sliding mode controller based on approach law and adaptive backstepping were designed, and optimal controller based on cerebellar model articulation controller (CMAC) was further designed. The tracking differentiator was added to eliminate the large chattering in the initial stage of the system. Finally, the simulation verification was carried out. It can be seen from the results that the control law designed has better control effects than traditional sliding mode control and can realize dynamic positioning of pod driven unmanned surface vessel under certain disturbances. INDEX TERMS Cerebellar model articulation controller, dynamic positioning system, pod propulsion, sliding mode control, unmanned surface vessel.