CAJ | 학술논문

为了解决三维环境中的无人机多路径规划问题，提出了一种基于改进概率地图的多目标蚁群算法。在构建地图时为了增加窄通道中的采样点数量，改进了概率地图法的采样策略，将落在威胁上的采样点移动到自由空间中，可以更好地覆盖规划环境。为了使蚁群算法可以得到多个解，提出了一种多目标蚁群算法。通过引入Pareto解集，播撒不同种类的信息素，使蚁群算法可以同时优化路径长度和威胁大小2个目标，并能得到一组非支配解，有利于决策者选择合适的路径。仿真结果表明，改进的概率地图法可以更好地覆盖规划环境，多目标蚁群算法可以得到一组解，并能收敛到最终解集。
위료해결삼유배경중적무인궤다로경규화문제，제출료일충기우개진개솔지도적다목표의군산법。재구건지도시위료증가착통도중적채양점수량，개진료개솔지도법적채양책략，장락재위협상적채양점이동도자유공간중，가이경호지복개규화배경。위료사의군산법가이득도다개해，제출료일충다목표의군산법。통과인입Pareto해집，파살불동충류적신식소，사의군산법가이동시우화로경장도화위협대소2개목표，병능득도일조비지배해，유리우결책자선택합괄적로경。방진결과표명，개진적개솔지도법가이경호지복개규화배경，다목표의군산법가이득도일조해，병능수렴도최종해집。
In order to solve the problems existing in the multi-path planning for UAV in the 3D environment, we improve the probability roadmap method ( PRM) by moving the sampling points to increase their number in their narrow passage so that the PRM can better serve the multi-path planning environment. Then we propose the multi-objective ant colony algorithm ( MACA) based on the PRM and apply it to the multi-path planning of the UAV. The MACA can optimize the path length and threat size of the UAV at the same time by updating their pheromones, thus obtaining a set of non-dominant solutions for the decision maker to select appropriate paths. To verify the effective-ness of the MACA, we simulate the multi-path planning environment as shown in Fig. 1; the simulation results, given in Fig.4 and 5 and Tables 1 and 2, and their analysis show preliminarily that our MACA based on PRM serves the 3D multi-path planning environment very well and can obtain a set of non-dominant solutions and con-verge to optimal solutions quickly.