Abstract: The leader-follower tracking problems of second-order multi-agent systems with intrinsic nonlinear dynamics are studied. It assumes that each following agent can access the relative position and velocity information with its neighbors, the position and velocity information of the leader is only accessed by a subset of the following agents, and the leader’s non-zero reference input cannot be available by any following agents. To track the active leader, a distributed adaptive consensus protocol is proposed for each following agent in the case that the interaction relationship among agents is undirected connected graph. The protocol effectively avoids the uncertainty of global information. The consensus tracking problem can be transformed into the stability problem of error system. Based on the theory of Lyapunov stability and matrix theory, it gets the sufficient conditions which guarantee the system to reach a leader-follower tracking consensus. Finally, a simulation example is given to verify the effectiveness of the obtained.

Key words: multi-agent systems, consensus, distributed control, adaptive control, leader

摘要： 基于带有非线性动态的二阶多智能体系统，研究了在有动态领导者条件下的跟踪一致性问题。假设跟随者只能获取邻居智能体的相对状态信息，只有一部分跟随者可以获得领导者的位置和速度信息，领导者的控制输入非零且不被任何一个跟随者可知。在通信拓扑为无向连通图的条件下，为了避免全局信息的不确定性，设计了分布式自适应控制协议。将系统的一致性问题转化为误差系统的一致性问题，通过Lyapunov稳定性理论和矩阵理论分析得到了该协议使系统达到一致的充分条件。最后用仿真例子证明了设计方法的有效性。

关键词: 多智能体系统, 一致性, 分布式控制, 自适应控制, 领导者