引用本文:陈志旺,夏顺,李建雄,王航,王昌蒙.考虑分配次序的无人机协同目标分配建模与遗传算法求解[J].控制理论与应用,2019,36(7):1072~1082.[点击复制]
CHEN Zhi-wang,XIA Shun,LI Jian-xiong,WANG Hang,WANG Chang-meng.Modeling of unmanned aerial vehicles cooperative target assignment with allocation order and its solving of genetic algorithm[J].Control Theory and Technology,2019,36(7):1072~1082.[点击复制]
考虑分配次序的无人机协同目标分配建模与遗传算法求解
Modeling of unmanned aerial vehicles cooperative target assignment with allocation order and its solving of genetic algorithm
摘要点击 2591  全文点击 937  投稿时间：2018-03-15  修订日期：2018-11-22
查看全文  查看/发表评论  下载PDF阅读器
DOI编号  10.7641/CTA.2018.80176
  2019,36(7):1072-1082
中文关键词  无人机  遗传算法  目标分配  分配模型
英文关键词  unmanned aerial vehicles  genetic algorithms  target assignment  assignment model
基金项目  国家自然科学基金
作者单位E-mail
陈志旺 燕山大学 czwaaron@ysu.edu.cn 
夏顺* 燕山大学 1203652746@qq.com 
李建雄 燕山大学  
王航 燕山大学  
王昌蒙 燕山大学  
中文摘要
      本文研究了动态战场环境中的多无人机协同目标分配(Multi-UAVs cooperative target assignment, MUCTA)问题, 首先通过分析UAV分配次序对打击任务总收益的影响, 设计了动态战场环境的更新规则, 将航程代价和任务代价作为惩罚项修正目标函数, 建立了考虑分配次序的UAVs协同目标分配优化模型. 然后针对模型的物理意义改进了遗传算法基因编码方式, 设计了MUCTA遗传算法, 该算法利用状态转移思想, 引进SDR算子获得多种分配次序种群, 同时以单行变异算子修正UAV与目标对应关系, 并采用最优个体法和轮盘赌法筛选子代个体. 最后仿真结果验证了所设计算法的有效性.
英文摘要
      This article is concerned with the Multi-UAVs cooperative target assignment(MUCTA) of dynamic battlefield environment. Firstly, By means of the influence of UAV allocation order on total revenue of strike task, the updating rules of dynamic battlefield environment are designed. The cost of flight path length and task are used as penalty term in objective function, and the optimization model of UAVs cooperative target assignment with allocation order is established. Secondly, the coding method of genetic algorithm is improved based on the physical significance of the optimization model, and the MUCTA genetic algorithm is proposed. According to state transition, SDR operator is used to obtain different population of various allocation order, single mutation operator is used to adjust the correspondence relation between UAVs and targets, the methods of Optimal individual selection and roulette are used to screen offspring individuals. Finally, simulation results verify the effectiveness of the algorithm.