Theta* algorithm is a searching-based path planning algorithm that gives an optimal path with more flexibility on route angle than A* method. The dynamics of USV is characterized by large inertia, so that larger turning angle is preferred. In view of the shortcomings of traditional Theta* algorithm, such as being hard to balance overall situation and details in long-distance planning, and lacking of waypoint replacement scheme when a waypoint is unreachable, it is inappropriate to make long-distance planning through Theta* algorithm directly. In order to ensure safe navigation for unmanned surface vehicle (USV), this paper proposes a multi-scale Theta* algorithm to solve these defects. Simulation result manifests the proposed scheme can provide a path clear of obstacles with several fold reduction in time consumption. The proposed planning method simplifies path planning process and contributes to the development of marine transportation.

Theta* 算法是一种基于搜索的路径规划算法，与 A* 方法相比，它给出了在路径角度上具有更大灵活性的最佳路径。USV的动力学特点是惯性大，因此首选较大的转弯角度。针对传统Theta*算法在远距离规划中难以兼顾全局和细节、航点不可达时缺乏航点替换方案等缺点，不宜通过Theta进行远距离规划* 直接算法。为了确保无人水面舰艇（USV）的安全导航，本文提出了一种多尺度Theta*算法来解决这些缺陷。仿真结果表明，所提出的方案可以提供一条无障碍路径，时间消耗减少了几倍。