The paper presents a ship motion planning method that adjusts the attractive and repulsive gain parameters of the artificial potential field (APF) algorithm using the deep deterministic policy gradient (DDPG) algorithm, to address the challenges of parameter design and limited adaptability of traditional APF algorithms. This approach leverages the speed advantage of the APF planning algorithm and the exploration capability of the DDPG reinforcement learning algorithm. By optimizing the parameters of APF with DDPG, the local optima problem can be alleviated in complex obstacle environments. Simulation results demonstrate that the proposed motion planning algorithm can generate reasonable trajectories and exhibit strong adaptability in complex obstacle scenarios.