Cooperative Control For Multi-agnet Systems And Its Applications To Distributed Spacecraft

The multi-agent system(MAS)is composed of a group of smaller and simpler agents that communicate each other through wireless links,such as distributed spacecraft system,unmanned aerial vehicle formation and so forth.Compared to the traditional agent,the advantages of MASs include strong fault-tolerant ability,high working efficiency,and low manufacturing cost.The cooperative control problem of MASs is a promising topic in control area,thus,it is necessary to investigate this problem.This thesis aims to investigate the cooperative control problem for multi-agent systems(MASs),then the proposed theoretical approaches are applied into distributed spacecraft systems.The content of this thesis includes the following four aspects:A modified order-reduction method for linear MASs is proposed,which can be utilized to reduce the numerical complexity of calculating linear matrix inequalities(LMIs).In MASs,the number of agents will affect the orders of LMIs,and the orders of LMIs will be very high if there are many agents in the whole network,which brings high computing cost.In this aspect,a modified order-reduction method is proposed,compared with the existing ones,the proposed method can highly reduce the computing time.Moreover,this order-reduction method is applied into the orbital transfer problem of distributed spacecraft.Considering the existence of external disturbances,an H?-based cooperative controller subject to energy-cost constraint is designed.It is shown that the orders of LMIs can be highly reduced with the help of the proposed order-reduction method.The cooperative tracking control problem of MASs under non-uniform time delays is studied.When multiple agents are communicating with each other,communication delays will exist because of the long distance and environment effect.In addition,communication delays should be different among different agents,and communication delays are called non-uniform in this case.In this aspect,two delay-dependent controllers are designed: integral sliding-mode controller and low-pass-filter-based dynamical sliding-mode controller.Simulation part applies the proposed method into the orbital transfer problem of distributed spacecraft to show its effectiveness and advantages compared to the H? controller.The fuzzy-based cooperative control technique for T?S fuzzy MASs is proposed.When using the traditional nonlinearity-feedback control method to deal with the cooperative control problem of nonlinear MASs,the existence of model uncertainties will affect the results.Moreover,the nonlinearity-feedback controller will highly increase the burden of actuators.In this aspect,we aim to construct the nonlinear MASs into T?S fuzzy MASs,and propose a fuzzy-based distributed controller.The theoretical method is applied into the spacecraft attitude coordination problem to show its effectiveness.Moreover,for the sake of comparison,the designed controller and a traditional nonlinearity-feedback controller are both provided in the simulation part.The sliding-mode control problem of nonlinear MASs is studied,and a chattering-free controller is designed.In practical applications,external disturbances will affect the performance of control systems,and the sliding-mode control is one of the most significant approaches to attenuate the disturbances.In this aspect,the nonlinear MAS is constructed into a fuzzy system,and a new model transformation method for fuzzy systems is proposed.Two fuzzy-based controllers including the integral sliding-mode controller and the dynamical sliding-mode controller are designed.To show the effectiveness of the proposed method,the simulation part applies the theoretical approach into the cooperative attitude tracking problem of distributed spacecraft.