水面无人艇冗余路径点约减方法研究——基于最优路径代价估计

针对复杂海洋环境下USV在线路径规划问题,提出一种顺序-随机两模式的冗余路径点约减方法。利用估计通过路径点的最优路径来制定其被约减的概率,并通过概率值的自适应变化考虑更多路径点的组合方案。采用顺序方法控制随机约减范围,快速发现潜在的优化路径。在约减过程中考虑海流等复杂海况对USV运动的影响,从而降低USV的碰撞概率。实验结果表明,本文方法可有效地简化并优化USV路径。

An USV Redundant Waypoints Pruning Method based on Optimal Path Cost Estimation

To solve the online USV path planning problem in complex ocean environment, we propose a two-mode redundant waypoints pruning method containing the sequential and probabilistic pruning modes. Regarding the problems of the path simplification as well as path optimization, an estimation method of the optimal path passing through a waypoint is raised to calculate the pruned probabilities of waypoints. At the same time, the pruned probabilities are changed adaptively to consider more waypoints combinations of a path to find a potentially better path. To reduce the pruning time, the scale of the probabilistic pruning process is constrained by the sequential pruning process. The influence caused by the complex sea state, such as current, is considered during the pruning process to decrease the collision probability of USV. The experimental results show that our method is better than the traditional methods in simplifying and optimizing USV paths.