Research On Robust Consensus Control For Multi-agent Systems Under Complex Conditions

Distributed coordinated control of multi-agent systems can make agents achieve an amazing overall effect,rather than simple collection of their individual functions,by interacting with each other,and it is the hot topic and frontier in the field of system control.Consensus problem is a basic problem of distributed coordinated control of multi-agent systems.In this thesis,for robust consensus control problems of multi-agent systems under complex conditions,the effects of time-delays,external disturbances and parameter uncertainties on consensus of multi-agent systems are analyzed,and the suitable control algorithms are designed to make multi-agent systems achieve consensus while satisfying the desired robust H_? performance index.Then these algorithms are applied to the formation control of unmanned aerial vehicles(UAVs).The main research contents and contributions of this thesis are as follows:For the robust rotating consensus control problems of the second-order multi-agent systems with parameter uncertainties,external disturbances and time-delay on the plane,sufficient conditions are proposed for achieving robust rotating consensus.Via system model transformations,the agreement and disagreement dynamics of the system are divided,a reduced-order system is obtained,and robust rotating consensus control problem of multi-agent system is transformed into a robust H_? control problem of this reduced-order system.Then,based on this reduced-order system,by employing H_? control theory,the sufficient conditions are derived for achieving robust rotating consensus of the second-order multi-agent systems on the plane.For the robust rotating consensus control problems of second-order multi-agent systems with parameter uncertainties,external disturbances and time-varying delay in three-dimensional space,and a low computational complexity algorithm is proposed for robust H_? rotating consensus control,then the optimization algorithms are proposed to improve the disturbance attenuation capacity and allowable delay upper bound.By variable substitution and matrix theory methods,the second-order multi-agent systems model is reduced-order and decoupled,and robust rotating consensus control problem can be transformed into robust H_? control problems of these decoupling subsystems.Then,based on these decoupling subsystems,by using Lyapunov-Krasovskii functionals and the convex property of Linear matrix inequalities(LMIs),a sufficient delay-dependent condition of lower complexity is derived for the realization of robust rotating consensus of the second-order multi-agent systems.Based on this condition,by using a cone complementarity linearization algorithm and an iterative algorithm,an H_? control algorithm is proposed for robust rotating consensus of the multi-agent systems.Extending this algorithm,the optimization algorithms are proposed to improve the disturbance attenuation capacity and allowable delay upper bound.For the robust consensus control problems of high-order multi-agent systems with parameter uncertainties,external disturbances and double time-varying delays,a lower conservative delay-dependent robust consensus sufficient condition is proposed for the high-order multi-agent systems,and a low computational complexity algorithm is proposed for robust H_? consensus control of high-order multi-agent systems,By variable substitution and matrix theory methods,the high-order multi-agent systems model is reduced-order and decoupled,and robust consensus control problem can be transformed into robust H_? control problems of these decoupling subsystems.For the case of double time-varying delays with nonuniform self-state and input delays,by applying some tools such as the Lyapunov-Krasovskii functionals,the Jensen inequalities and the Newton-Leibniz formulas with free weight matrices,a lower conservative delay-dependent robust consensus sufficient condition is derived.Based on this condition,an H_? control algorithm is proposed for robust consensus of the multi-agent systems.Extending this algorithm,the optimization algorithms are proposed to improve the disturbance attenuation capacity and allowable delay upper bound.Based on the theoretical results of robust consensus control for multi-agent system in this thesis,the controller synthesis rules and algorithms are proposed to realize time-varying formation of UAVs swarm.The nonlinear model of UAV is transformed into a linear double-integrator model by employing the feedback linearization method.Then,a distributed control protocol based on the state feedback of UAV neighbors is designed.By the variable substitutions,robust time-varying formation control problem of UAVs swarm is transformed into the robust consensus control problem of multi-agent systems.Finally,based on the results of robust consensus control for multi-agent system in this thesis,the controller synthesis criterions and algorithms are proposed to realize robust time-varying formation control of UAVs swarm.