带动态领导者的一类多智能体系统跟踪控制  

Tracking control for multi-agent systems with active leader

作　　者：冯元珍[1,2] 徐胜元[1] 苗国英[1] 

机构地区：[1]南京理工大学自动化学院,江苏南京210094 [2]南京人口管理干部学院基础部,江苏南京210042

出　　处：《南京理工大学学报》2013年第3期356-359,共4页Journal of Nanjing University of Science and Technology

基　　金：教育部高等学校博士学科点专项科研基金(20113219110026)

摘　　要：为降低保守性,该文分别在固定和切换有向拓扑结构下,研究了带动态领导者的一类多智能体系统的跟踪控制问题。基于近邻居控制算法及构造状态观测器实现跟踪控制。采用参数相关的Lyapunov函数,在固定及切换有向拓扑结构下,当领导者的加速信息未知时,给出了实现跟踪控制的充分条件及跟踪误差的上界;在固定拓扑结构下,当领导者的加速信息已知时,给出了实现跟踪控制的充要条件。数值算例验证了相关结论的有效性。To reduce the conservativeness,the tracking control problems for a class of multi-agent systems with an active leader are considered when the directed topologies are fixed and switching respectively.Nearest-neighbor based control law is proposed and state observer is constructed to realize tracking control.In both fixed and switching cases,by using parameter-dependent Lyapunov functions,sufficient conditions are derived for realizing tracking control and upper bounds of tracking errors are estimated when the acceleration information of the leader is unknown.For the case of fixed topology,necessary and sufficient conditions are given when the acceleration information of the leader is known.Numerical examples demonstrate the effectiveness of relevant results.

关 键 词：动态领导者 多智能体系统 跟踪控制 LYAPUNOV函数 有向拓扑 

分 类 号：TP13[自动化与计算机技术—控制理论与控制工程]