Uncertain Multiagent Systems With Integrated Adaptive Control Gains: Neuro-Adaptive Asymptotic Consensus Tracking
Abstract
Despite advancements in the field of consensus control for multiagent systems (MASs), achieving asymptotic consensus in a fully distributed manner remains a formidable challenge, particularly for systems with unknown nonlinear dynamics. In this context, this article considers the neuro-adaptive asymptotic consensus tracking problem of MASs subject to functional uncertainties under directed communication topology. A set of smooth neuro-adaptive fully distributed controllers with new kinds of adaptive update laws is proposed, where the updated laws are designed to adaptively learn the integrated control gains, neural network weights, coupling strengths, and time-varying boundaries, respectively. Employing the Lyapunov stability theory, it has been demonstrated that the asymptotic consensus tracking can be reached if the communication topology graph includes a spanning tree. Finally, the theoretical result is validated through performing simulation on the multiple autonomous aerial vehicles model.