Smooth Path Planning for Unmanned Surface Vehicle in Environments with Unknown Polygonal Obstacles

Path optimization is a critical issue in unmanned surface vehicle (USV) path planning. The smoothness of the generated path and the number of turning points directly affect the ability of the USV to navigate safely. To address the global path planning problem in unknown polygonal obstacle environments, a novel path planning method with obstacle avoidance based on visibility graph is presented in this paper. The polygonal obstacles with different shape are modeled to get a smooth, feasible and collision-free path. The USV only needs to consider the closest obstacle in view and choose the shortest path to the target from the visibility tangent graph formed by the three nodes (its current position, the nearest obstacle, and the target). Upon encountering an obstacle, the USV navigates along the obstacle boundary and transitions to the next tangent path toward the target. The method can effectively reduce the computational complexity and generate a smooth path. The simulation results are consistent with the theoretical values, which verifies the feasibility of the proposed method.