有领导者线性多智能体系统一致性的分析与设计

Consensus analysis and design problem for leader-following linear multi-agent systems

研究有向信息拓扑下有领导者线性多智能体系统的一致性分析与设计问题。利用提出的线性变换,将领导者有扰动输入的多智能体系统一致性问题转换为输入到状态稳定性问题。得到有向信息拓扑下有领导者线性多智能体系统达到一致的基于矩阵Hurwitz稳定的判据和误差估计函数,同时设计反馈增益矩阵,并将有领导者多智能体系统的一致性问题扩展到编队控制问题。数值实例验证表明:所得理论具有有效性。

Consensus analysis and design problem for leader-following linear multi-agent systems（LLMASs） with directed information topology was investigated. A proper linear transformation was proposed to transform the consensus problem of a leader with disturbance inputs to the input-to-state stability problem of a corresponding linear system. Then, a new consensus criterion in terms of Hurwitz stability of matrices was given for LLMASs achieving consensus with directed information topologies, and the tracking error was estimated. Moreover, a design process to determine the feedback gain matrix in the consensus protocol was proposed. Finally, the consensus was extended to the formation control. Numerical examples are given to validate the above theoretical results.