基于置信椭圆的无人艇区域覆盖搜寻规划优化

为解决水上遇险目标搜寻存在搜寻区域过大、搜寻代价过高等问题, 提出一种基于置信椭圆的无人艇区域覆盖搜寻规划方法。首先, 利用高斯混合模型聚类算法划分搜寻区域, 再基于腐蚀膨胀的置信椭圆确定最佳搜寻区域边界, 实现目标包含概率和单位面积粒子数全局最优。然后, 构造适应椭圆搜寻区域边界特征的无人艇转向模型, 优化非工作路径。最后, 以搜寻探测概率和总路径为优化目标, 采用带精英策略的非支配排序的遗传算法(nondominated sorting genetic algorithm II, NSGA-II)优化得到最佳搜寻规划路径, 实现无人艇高效区域搜寻规划。与常规矩形区域覆盖搜寻规划对比, 能够在达到相同搜寻成功率下显著减少搜寻代价投入。

Optimization of USV area coverage path planning based on confidence ellipsoid

To solve the issues of too-large search area and too-high search cost for distressed targets overboard, an unmanned surface vehicle(USV) area coverage path planning method based on confidence ellipsoid is proposed. Firstly, a Gaussian mixture model clustering algorithm is used to divide the search area. And then the optimal search area boundary is determined based on the confidence ellipsoid of corrosion expansion to achieve the global optimum of target containing probability and the number of particles per unit area. Then, the USVs steering model is built to adapt to the boundary characteristics of the ellipsoid search region, which could optimize the non-working distance. Finally, the nondominated sorting genetic algorithm II (NSGA-II) is used to optimize the optimal path planning with the target detection probability and total distance as the optimization objectives, which realizes the efficient area coverage path planning for USVs. Compared with conventional rectangular area coverage path planning methods, it can significantly reduce the search effort cost under the same successful searching probability.