基于改进型遗传算法的多无人机任务分配  

Multi⁃UAV task assignment based on improved genetic algorithm

作　　者：徐宾王 宁芊[1] Xu Binwang;Ning Qian(School of Electronic Information,Sichuan University,Chengdu 610000,China)

机构地区：[1]四川大学电子信息学院,成都610000

出　　处：《现代计算机》2023年第5期15-21,共7页Modern Computer

摘　　要：无人机集群的决策过程中存在个体任务分配问题,如在无人机集群执行侦察任务时存在的任务点分配问题,针对解决该问题时存在的计算时间长,搜索性能差等问题,提出一种针对算法机制和并发方式的改进方法。对无人机任务分配问题建立模型,将其抽象为多旅行商问题,通过遗传算法来解决,并将无人机实际飞行场景中的代价设置为动态和静态两种情况。在此基础上,通过改变计算适应度并行方式和添加缓存命中机制以减少算法冗余计算,提升算法适应度计算效率,采取优化的遗传算法淘汰机制使算法收敛速度提升。仿真结果验证,淘汰机制和并行缓存的优化使遗传算法具有更好的寻优能力和收敛速度,能够在较短的时间内规划出代价较小的航迹。There are individual task assignment problems in the decision⁃making process of UAV swarms,such as the task point assignment problem when UAV swarms perform reconnaissance tasks.Based on this,an improved method for algorithm mechanism and concurrency is proposed.A model is established for the task assignment problem of UAV,which is abstracted as a multi⁃travel salesman problem,and the problem is solved by genetic algorithm.On this basis,by changing the parallel method of calculating fitness and adding a cache hit mechanism to reduce redundant calculation of the algorithm,improve the efficiency of al⁃gorithm fitness calculation,and adopt an optimized genetic algorithm elimination mechanism to improve the algorithm convergence speed.The simulation results verify that the optimization of the elimination mechanism and the parallel cache enables the genetic algorithm to have better optimization ability and convergence speed,and can plan the trajectory with less cost in a short time.

关 键 词：任务分配 并行方式 缓存命中 淘汰机制 

分 类 号：V279[航空宇航科学与技术—飞行器设计] TP18[自动化与计算机技术—控制理论与控制工程]