基于自适应蚁群算法的无人飞行器航迹规划  

Unmanned aerial vehicle flight path planning based on adaptive ant colony optimization algorithm

作　　者：胡中华[1,2] 赵敏[1] 刘世豪[3] 章婷[3] 

机构地区：[1]南京航空航天大学自动化学院,江苏南京210016 [2]中国电子科技集团第38研究所,安徽合肥230031 [3]南京航空航天大学CIMS工程技术研究中心,江苏南京210016

出　　处：《计算机集成制造系统》2012年第3期560-565,共6页Computer Integrated Manufacturing Systems

基　　金：航空科学基金资助项目(2009ZC52041);国家自然科学基金资助项目(60974105;51005121);南京航空航天大学基本科研业务费专项资助项目(NP2011005)~~

摘　　要：为求解无人飞行器航迹规划问题,提出自适应蚁群算法,区别于标准蚁群算法的全部搜索模式,该算法采用局部搜索模式。首先根据起始节点与目标节点的相对位置关系选择相应的搜索模式,然后计算各个待选节点的转移概率,最后按照轮盘赌规则选择下一个节点。仿真结果表明,自适应蚁群算法具有搜寻节点数少、速度快等优点,在降低了航迹代价的同时,减小了计算时间。此外,自适应蚁群算法可以避免奇异航迹段的出现,从而保证所获的航迹实际可飞,表明所提算法整体性能明显较标准蚁群算法优异。To solve the problem of Unmanned Aerial Vehicle（UAV）route planning, Adaptive Ant Golony Optlmiza- tion（AACO）algorithm was proposed. Different from the global search mode of standard Ant Colony Optimizatio （ACO）, local search mode was adopted by AACO. Based on the relative position of starting node and destination node, one of the appropriate search mode in four was selected, and transition probabilities of each candidate node were calculated. The next node was selected according to the roulette principle. The simulation result showed that AACO algorithm had advantages such as few search nodes, quick speed and so on. It could reduce flight path cost and computing time. In addition, AACO could also avoided singular flight path segment, thus the attained practical flight path could fly was guaranteed. Therefore,the performance of AACO was much better than standard ACO.

关 键 词：无人飞行器 航迹规划 蚁群优化算法 自适应搜索 航迹代价 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程] TP301[自动化与计算机技术—控制科学与工程]