Abstract: [Objectives] This paper presents a tracking control method based on dynamic programming guidance to address the autonomous recovery challenge of underactuated unmanned surface vehicles (USVs). [Methods] At the kinematic level, the parallel approaching guidance is combined with the dynamic window algorithm to guide the USV in achieving target tracking and dynamic obstacle avoidance. At the dynamic level, considering the uncertainties in the model and recovery environment, a radial basis function neural network is employed to design a dynamic sliding mode controller for tracking control of the guidance output. Finally, the stability of the system is analyzed using Lyapunov theory. [Results] Simulation results demonstrate that the proposed method enables the USV to exhibit stable tracking performance, effectively avoid dynamic obstacles during the recovery process, and adapt to uncertain factors in the model and unknown environmental disturbances. [Conclusions] The proposed method exhibits strong robustness and flexibility, providing a reference for guidance and target tracking of USVs during recovery in dynamic environments.