领导者-跟随者多智能体系统的分布式协同控制  

Distributed Collaborative Control of Leader-follower Multi-agent Systems

作　　者：张星红 李齐齐 陈慧波 郑秀丽 丁昊 ZHANG Xinghong;LI Qiqi;CHEN Huibo;ZHENG Xiuli;DING Hao(School of Electrical Engineering&Automation,Henan Institute of Technology,Xinxiang 453003,China)

机构地区：[1]河南工学院电气工程与自动化学院,河南新乡453003

出　　处：《河南工学院学报》2025年第1期18-22,共5页Journal of Henan Institute of Technology

基　　金：国家自然科学基金(62076028);河南省高等学校重点科研项目(22A413005);河南省科技攻关项目(212102210247)。

摘　　要：针对多智能体系统中的未知参数,采用具有惩罚因子的改进投影算法进行辨识。基于参数估计更新律和参考信号,设计了领导者的局部自适应控制律;基于参数估计更新律和邻居的历史信息,设计了每个跟随者的局部自适应控制律。在设计的分布式自适应控制律下,领导者-跟随者多智能体系统具有下列控制性能:(1)领导者跟踪参考信号;(2)每个跟随者跟踪领导者;(3)所有智能体跟踪参考信号,实现了多智能体系统的分布式协同控制。最后,通过仿真结果验证了理论的有效性。For the leader-follower multi-agent systems with unknown parameters,the improved projection algorithm with the penalty factor is used to identify the parameters.For the hidden leader,a local adaptive control law is designed based on the parameter estimation update law and the reference signals.In the face of situations where each follower has different neighbors with different degrees of influence,a local adaptive control law is designed based on the parameter estimation update law and the historical neighbors’information.Under the designed distributed adaptive control laws,the multi-agent systems with the fixed directed topology have the following control performance:(1)the hidden leader tracks the reference signals;(2)each follower follows the leader;(3)all agents track the reference signals,and the whole system achieves distributed collaborative control.Finally,the effectiveness of the given theory have been verified by the digital simulation results.

关 键 词：协同控制 分布式控制 多智能体系统 领导者 未知参数 

分 类 号：TP273.2[自动化与计算机技术—检测技术与自动化装置]