The problem of autonomous obstacle avoidance of unmanned surface vehicles (USV) under dense obstacle sea conditions is studied.

An improved dual-window dynamic window approach (DWA) obstacle avoidance algorithm based on a fuzzy control strategy is proposed; that is, a sensing window based on an onboard sensor is designed on the basis of the conventional speed window, and a dual-window model composed of the speed window and sensing window is further optimized. The speed space is constrained and the weight of the evaluation function is dynamically adjusted based on the fuzzy control strategy in accordance with the obstacle distribution state and distance between the USV and obstacles.

The simulation experimental results show that, compared with the original DWA algorithm, the path planned by the improved algorithm under unknown dense obstacle sea conditions is smoother and more reasonable, which not only solves the problem of USVs detouring outside dense obstacle groups, but also improves the safety of obstacle avoidance navigation and reduces the number of iterations and running time by more than 20%.

The results of this study have certain reference value for research on the autonomous obstacle avoidance technology of USVs.