一阶时滞多智能体系统一致性研究  

Research on Consistency of Two Order Delayed Multi-agent Systems with Directed Topology

作　　者：朱雪芳 彭力[2] ZHU Xuefang;PENG Li(Jiangsu Union Technical Institute,Wuxi 214028;Engineering Research Center of Internet of Things Application Technology of the Ministry of Education (Jiangnan University),Wuxi 214122)

机构地区：[1]江苏联合职业技术学院无锡机电分院,无锡214028 [2]物联网应用技术教育部工程研究中心(江南大学),无锡214122

出　　处：《现代制造技术与装备》2019年第8期37-42,44,共7页Modern Manufacturing Technology and Equipment

基　　金：江苏省高校自然科学研究项目（16KJB510044）

摘　　要：针对时滞多智能体系统,通过深入分析研究传统一致性算法,为解决传统一致性协议不能同时满足较高时滞鲁棒性及一致性收敛速度的问题,将加权平均预测机制引入传统的一致性协议中,并应用到带无向拓扑的一阶时滞多智能体系统,获得保证其收敛条件,并证明在已有一致性协议中引入适当长度加权平均预测可以提高系统时滞鲁棒性,加快实现一致的收敛速度。For the multi-agent system with time-delay, the traditional consistency algorithm is deeply analyzed. To solve the problem that the traditional consistency protocol can not satisfy the high time-delay robustness and the consistency convergence speed, the weighted average prediction mechanism is introduced into the traditional consistency. In the sexual agreement, and applied to the first-order time-delay multi-agent system with undirected topology, the convergence condition is guaranteed, and it is proved that the introduction of appropriate length weighted average prediction in the existing consistency protocol can improve the system delay robustness., speed up to achieve consistent convergence speed.

关 键 词：多智能体系统 一致性 收敛速度 鲁棒性 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]