任意有限维空间中群体行为的研究

Research on swarm behavior in arbitrarily finite dimensional spaces

探讨了任意有限维空间中的智能群体的聚集行为和迁徙行为,提出了一类基于局部信息的群体行为控制方法。从虚拟相互作用力的角度理解分析群体行为,采用吸引/排斥力的数学方法实现个体间的相互作用,并利用求解任意有限维空间点集的凸壳方法,将二维空间中的群体行为模型推广到任意有限维空间中,实现了一类基于局部信息的且适合于任意有限维空间的群体行为控制。在此基础上分析了系统的稳定性,并通过仿真实验验证了该方法的有效性。

Collective motions that consist of aggregation behavior and flocking behavior in n-dimensional（n≥2） spaces are studied for intelligent swarm.Based on individual local information a class of control laws for swarm behavior is proposed.The collective motion is analyzed with virtual force that is implemented by an attraction/repulsion function.By introduction of a convex hull method,a collective motion control for a two-dimensional swarm is extended to n-dimensional swarm,and a distributed collective motion control for intelligent swarm in n-dimensional spaces based on local information is proposed.The stability properties of the swarm behavior are analyzed.Simulation results are presented to demonstrate the validity of the models.