期刊
IEEE TRANSACTIONS ON AUTOMATIC CONTROL
卷 66, 期 11, 页码 5612-5618出版社
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/TAC.2021.3056336
关键词
Symmetric matrices; Convergence; Adaptive learning; Adaptive control; Robot kinematics; Protocols; Multi-agent systems; Containment control; cooperative adaptive learning control; cooperative finite-time excitation
资金
- National Science Foundation [CMMI-1952862]
This article addresses the problem of cooperative adaptive containment control for multiagent systems through a novel control architecture that achieves both containment control and parameter adaptive learning simultaneously. The proposed method achieves exponential convergence of containment tracking errors to zero and adaptation parameters to their true values under a mild cooperative finite-time excitation condition.
This article addresses the problem of cooperative adaptive containment control for multiagent systems, which specifies the objective of jointly achieving containment control and accurate adaptive learning/identification of unknown system parameters. We consider a class of linear uncertain multiagent systems with multiple leaders subject to bounded unmeasurable inputs and multiple followers subject to unknown system dynamics. A novel cooperative adaptive containment control architecture is proposed, which consists of a discontinuous nonlinear state-feedback control law and a filter-based cooperative adaptation law. This new control architecture is compelling in the sense that exponential convergence of both containment tracking errors to zero and adaptation parameters to their true values can be achieved simultaneously under a mild cooperative finite-time excitation condition. This condition significantly relaxes existing ones (e.g., persistent excitation and finite-time excitation) for parameter identification in adaptive control systems. Effectiveness of the proposed approach has been demonstrated through both rigorous analysis and a case study.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据