针对多障碍陆战场路径规划的改进A*算法研究

An Improved A* Algorithm for Path Planning in Multi-Obstacle Land Battlefield

传统A*算法在环境障碍类型多,道路特性复杂的陆战场进行路径规划时,求解所得路径并非最短且转角偏大,同时还存在所得路径实际可能无法通过的不足之处。本实验采用改进A*算法解决多障碍避让问题,求解最佳路径。首先结合元胞自动机理论,将搜索数据结构改进为扩展Moore型,其次改进估价函数计算方式,添加多组适应函数,归纳判定条件,使搜索邻域可直连第二层节点,降低了最小转角及路径长度。最后根据狭隘路段通行条件,再次改进搜索方式,添加二次搜索函数,从而智能识别并绕开狭隘路口,同时还留出了一定绕出空间及安全距离。经Matlab2018路径搜索仿真表明,该改进A*算法相较于传统A*算法在复杂路径规划问题上具有一定的先进性与优越性。

When used for path planning in the land battlefield with many different types of environmental obstacles and complex roadway characteristics,the path obtained by the traditional A*algorithm is not the shortest.It also has large corner and may be impassable in reality.In this experiment,an improved A*algorithm was proposed to help avoid multiple obstacles and solve the optimal path.Based on the concept of cellular automata,the search data structure was first changed to an extended Moore type.To reduce the minimum corner and path length,the search neighborhood was directly connected to the nodes in the second layer by improving the calculation method of evaluation function,adding multiple sets of fitness functions,and summarizing the judging criteria.The search method was further improved by adding a secondary search function based on the traffic conditions of narrow road sections,so that the algorithm can intelligently identify and bypass narrow intersections,while reserving a certain amount of space and a safe distance for bypassing.The simulation based on Matlab2018 path search shows that compared with the traditional A*algorithm,the improved A*algorithm is more advanced and superior in complex path planning.