基于RQPSO-DMPC的多无人机编队自主重构控制方法  

Autonomous reconfiguration control method for multi-UAV formation based on RQPSO-DMPC

作　　者：周绍磊[1] 康宇航 史贤俊[1] 戴邵武[1] 周超[3] 

机构地区：[1]海军航空工程学院控制工程系 [2]91899部队 [3]92857部队

出　　处：《北京航空航天大学学报》2017年第10期1960-1971,共12页Journal of Beijing University of Aeronautics and Astronautics

基　　金：国家自然科学基金(61102167);航空科学基金(20135184007)~~

摘　　要：针对敌方防御区域内各种威胁,为了实现隐蔽突防并实施对敌有效打击,在突防过程中多无人机(UAV)编队需要进行重构控制,并且编队内的相互避碰问题与通信约束问题也需考虑。通过建立无人机虚拟领航编队模型并引入邻居集,采用分布式模型预测控制(DMPC)同时构建多无人机编队的重构代价函数,提出采用改进量子粒子群优化(RQPSO)算法进行求解,并将求解结果与采用粒子群优化算法的结果进行对比。仿真结果表明,本文算法能够有效控制多无人机编队完成自主重构,实现安全隐蔽突防任务。For various threats in the enemy defense area, in order to achieve covert penetration and im- plement effective combat against enemy, the unmanned aerial vehicle （UAV） formation needs to be reconfig- ured in the process of penetration, and the muhi-UAV collision avoidance problem and communication con- straint problem within the formation also need be considered. By establishing the virtual leader formation mod- el and introducing the neighbor set, this paper adopts distributed model predictive control （DMPC） , recon- structs the cost function of muhi-UAV formation reconfiguration, and proposes that the cost function is solved by adopting the revised quantum-behaved particle swarm algorithm. The solving result is compared with the re- sult obtained by particle swarm algorithm. Simulation result shows that this algorithm can control multi-UAV formation＇ autonomous reconfiguration effectively and achieve covert penetration safely.

关 键 词：无人机(UAV) 编队重构 邻居集 分布式模型预测控制(DMPC) 量子粒子群 

分 类 号：V249.122[航空宇航科学与技术—飞行器设计] TP391[自动化与计算机技术—计算机应用技术]