摘要
为更好解决复杂环境的路径规划问题,研究在高程信息、地形坡度、地表类型等多约束条件影响下的特种无人车多目标A^(*)算法。将已知环境信息分类建成不同信息层栅格地图,叠加后形成2.5维融合栅格地图;根据不同约束条件建立路径多目标优化函数,并根据优化目标改进A^(*)算法的代价函数;采用熵值法对改进A^(*)算法得到的多条路径进行综合评价,筛选多目标优化效果最佳的路径;仿真结果表明在模拟的复杂环境下,改进的A^(*)算法规划的路径在长度、平稳性、无人车行驶时间、隐蔽性等方面都能够达到优化效果,验证了在复杂地形约束下,该改进算法对无人车路径多目标优化的可行性和有效性。
In order to better solve the path planning problem in complex environment,the multi-objective A^(*)algorithm of special unmanned vehicle under the influence of multi constraints such as elevation information,terrain slope and surface type is studied.The known environmental information was classified and established into grid maps of different information layers,which were superimposed to form a 2.5-dimensional fusion grid map.The path multi-objective optimization function was established according to different constraints,and the cost function of A^(*)algorithm was improved according to the optimization objective.The entropy method was adopted to comprehensively evaluate multiple paths obtained by the improved A^(*)algorithm,screening out the path with the best multi-objective optimization effect.The simulation results show that in the simulated complex environment,the path planned by the improved A^(*)algorithm can achieve the optimization effect in terms of length,stability,unmanned vehicle driving time and concealment.It is verified that the improved algorithm is feasible for the multi-objective optimization of unmanned vehicle path under the constraint of complex terrain.
作者
刘健
沈芸亦
邱锦
罗亚松
Liu Jian;Shen Yunyi;Qiu Jin;Luo Yasong(College of Weapon Engineering,Naval University of Engineering,Wuhan 430033,Hubei,China)
出处
《计算机应用与软件》
北大核心
2025年第8期297-305,381,共10页
Computer Applications and Software
基金
全军军事类研究资助课题(JY2020B117)。
