多无人海洋运载器协同编队任务分配方法

    Coordinated task allocation for UMV formation control

    • 摘要: 为更好地处理海面溢油,针对无人海洋运载器集群队形控制过程中任务分配延迟、收敛速度慢和易陷入局部最优的问题,提出将上层的巡回规划过程与下层的任务指派过程紧密结合的多无人海洋运载器协同编队任务分配方法。首先以匈牙利算法建立多运载器代价矩阵并完成运载器集群个体“可能围油任务”的协同分配,其次以蚁群算法求解运载器集群航迹规划问题,最后得到运载器集群最短闭循环巡回路径。仿真结果表明,通过该方法求得的总航迹长度缩短了15.13%,算法平均运行时间减少了46.57%,任务分配合理性得以有效提升。

       

      Abstract: The control of unmanned marine vehicle(UMV) cluster for treatment of oil spill at sea is studied. The lag of task allocation, low convergence speed and possibility to be trapped in local optimal solution are found quit often in the process of formation control. A new control design is proposed where, tour route planning, the upper layer process, and task allocation, the lower layer process, are tightly combined. The new design uses Hungarian algorithm to build the multi-UMV cost matrix and allocate “tentative tasks” to individual UMVs in the oil containment action. The ant colony optimization is used to solve the problem of tour route planning and generate the shortest closed loop of touring for the UMV cluster. Simulation is performed, which shows that the travel distance reduction of 15.13% and average computing time reduction of 46.57% are achieved with the new design.

       

    /

    返回文章
    返回