基于虚拟领航者的智能群体群集运动控制  

Virtual leaders-based control of flocking motion of intelligent swarm

作　　者：王冬梅[1] 方华京[1] 

机构地区：[1]华中科技大学控制科学与工程系,湖北武汉430074

出　　处：《华中科技大学学报（自然科学版）》2008年第10期5-7,共3页Journal of Huazhong University of Science and Technology(Natural Science Edition)

基　　金：国家自然科学基金资助项目(60574088;60274014)

摘　　要：通过借鉴预测控制的基本思想,研究了智能群体的群集运动,并建立了一种个体在运动过程中能根据周围环境自适应选取虚拟领航者的群集运动控制模型.该模型描述了生物群集过程中无领导者的自然现象,克服了传统leader-follower模式的单向信息流的缺陷,增强了系统的自适应性与鲁棒性,提高了系统的稳定性.仿真结果表明:与已有文献相比,该模型收敛速度较快;形成紧凑的编队后,队形稳定运行的速度也较快.Flocking motion of intelligent swarm was studied by the mythology of predictive control. A flocking control model was built, in which an individual could choose virtual leaders adaptively from surroundings during its motions. The model described the phenomenon in nature ＇there no leader in flocking motion of swarm＇, which overcame the limitation of leader to follower information flow, and improved the self-adaptability, robustness and stability of the system. Simulation shows that the pro posed model improves the convergence speed and the stable moving speed after group into a tight formarion.

关 键 词：智能群体 协调控制 虚拟领航者 稳定性 群集运动 

分 类 号：TP13[自动化与计算机技术—控制理论与控制工程]