Aiming at the problems of insufficient mobility and low trajectory tracking accuracy of traditional buoys, A new type of surface mobile buoy was designed, and a trajectory tracking algorithm integrating the A* algorithm and nonlinear model predictive control (NMPC) was also designed. First of all, establish the mathematical model of the three-degree-of-freedom surface mobile buoy. Secondly, the A*-NMPC algorithm is designed. By combining the global path planning ability of the A* algorithm with the dynamic optimization characteristics of NMPC, a collaborative control framework is formed. Finally, through conducting simulation experiments and comparing and analyzing the trajectory tracking effect with that of linear model predictive control (MPC). The results show that A*-NMPC exhibits significant advantages in the scenario of sudden course changes. The maximum longitudinal error is reduced by 50% compared with MPC, the lateral error is reduced by 80%, and it has the capabilities of dynamic obstacle avoidance and rapid response.