Neuro-Adaptive Consensus Tracking of Multiagent Systems With a High-Dimensional Leader
研究了有向通信拓扑下不确定多智能体系统对高维领导者的分布式一致性跟踪问题，为每个跟随者设计了分布式鲁棒自适应神经网络控制器和局部观测器，确保状态同步并允许拓扑切换。
This paper is concerned with the distributed consensus tracking problem of uncertain multiagent systems with directed communication topology and a single high-dimensional leader. Compared with existing related works, the dynamics of each follower in the present framework are subject to unmodeled dynamics and unknown external disturbances, which is more practical in various applications. Furthermore, the dimensions of leader's dynamics may be different with those of the followers' dynamics. Under the mild assumption that each follower can directly or indirectly sense the output information of the leader, a distributed robust adaptive neural network controller together with a local observer are designed to each follower to ensure that the states of each follower ultimately synchronize to the leader's output with bounded residual errors under a fixed topology. By appropriately constructing some multiple Lyapunov functions, the derived results are further extended to consensus tracking with switching directed communication topologies. The effectiveness of the analytical results is demonstrated via numerical simulations.