搜索

x
中国物理学会期刊

时延多智能体系统分组一致性分析

CSTR: 32037.14.aps.61.220202

Group consensus analysis of multi-agent systems with delays

CSTR: 32037.14.aps.61.220202
PDF
导出引用
  • 研究了拓扑结构为连通无向图与连通二分图的一阶时延多智能系统分别在两类控制协议下的分组一致性问题. 基于广义Nyquist准则与频域控制理论的方法, 分析并得到了多智能体系统渐进分组收敛一致的充分条件. 通过该条件发现系统分组一致性的达到与系统的时延以及智能体间的耦合强度相关.同时, 时延的大小影响着系统的收敛速度与动态性能. 仿真实验的结果进一步验证了理论分析所得结论的正确性.

     

    In this paper, a new consensus problem is investigated which is group consensus. It contains such consensus problem as a special case that all agents in a network reach a consistent value asymptotically. Consider a dynamic multi-agent system with connected undirected or connected bipartite graph topology and delays, two novel consensus protocols are introduced to solve the group consensus problem. The convergence analysis is discussed and the sufficient condition for system group convergence is obtained based on the frequency-domain analysis and generalized Nyquist criterion, respectively. And we know that the condition of group convergence is only depended on system's time delay and the adjacent weight between the agents, meanwhile, delay can affect the dynamic characteristics of the system. Finally, simulations are provided to demonstrate the effectiveness of our theoretical results.

     

    目录

    /

    返回文章
    返回