Predictor-Based Extended-State-Observer Design for Consensus of MASs With Delays and Disturbances
针对输入输出均存在时延和外部扰动的多智能体系统，提出一种基于预测器的扩展状态观测器，实现领导者-跟随者一致性控制，且控制器不含过去控制输入的积分项，易于实现。
In this paper, we study output feedback leader-follower consensus problem for multiagent systems subject to external disturbances and time delays in both input and output. First, we consider the linear case and a novel predictor-based extended state observer is designed for each follower with relative output information of the neighboring agents. Then, leader-follower consensus protocols are proposed which can compensate the delays and disturbances efficiently. In particular, the proposed observer and controller do not contain any integral term of the past control input and hence are easy to implement. Consensus analysis is put in the framework of Lyapunov-Krasovskii functionals and sufficient conditions are derived to guarantee that the consensus errors converge to zero asymptotically. Then, the results are extended to nonlinear multiagent systems with nonlinear disturbances. Finally, the validity of the proposed design is demonstrated through a numerical example of network-connected unmanned aerial vehicles.