Cooperative Output Regulation For Linear Uncertain Multi-agent Systems

In recent years,the cooperative output regulation problem of multi-agent syst,ems has attracted more and more attention in the field of control due to its practical characteristics.The output regulation problem of multi-agent systems can be described as:given a set of uncertain multi-agent systems and a connected network topology,design a distributed controller so that multi-agent systems can achieve asymptotic tracking as well as suppress a disturbance signal.This thesis considers robust control and adaptive control,respectively,to realize cooperative output regulation problem for linear uncertain multi-agent system,which can be summarized as:1.We will use robust control method to solve the output regulation problem of multi-agent systems.For single-input single-output system,we will show that the common controller for fixed network is also efficient for the more general uniformly connected switching network.In contrast to the tradi-tional quadratic Lyapunov function in literatures,the Lipchitz continuous Lyapunov function is developed here for the closed-loop system to guaran-tee its internal stability.For multi-input multi-output system,we further consider the output regulation problem for a class of general linear uncer-tain multi-agent systems subject to a well-defined vector relative degree.By proposing some suitable internal model,the problem is first converted into the auxiliary cooperative stabilization problem of the augmented sys-tem in the so-called strict feedback normal form.This auxiliary problem is then solved by some developed robust control techniques,such as multiple high-gain feedback and mixed distributed observer,leading to an effective distributed output feedback regulator synthesis for the original cooperative output regulation problem.2.We will use adaptive control method to solve the output regulation problem of multi-agent systems.In this thesis,an adaptive output feedback controller is designed to realize the asymptotic tracking and disturbance rejection of multi-agent systems.The advantage of this method is that we relax the limitations of the robust control method in which the uncertain parameters belong to the compact set,and further discuss the case where the uncertain parameters belong to unknown set.Based on internal model principle,we will first convert the output regulation problem into the stabilization problem of augmented systems.Nonlinear stabilization controller and adaptive law are then provided by recursive iterative method to achieve internal stability of the closed-loop system.Finally,a distributed nonlinear output feedback controller for cooperative output regulation of the original system is obtained by inverse coordinate transformation.For single-input single-output system,we consider a class of linear multi systems subject to nonidentical relative degrees.For the multi-input multi-output system,we consider a class of multi-agent systems which satisfies strict relative degree condition.