State And Output Feedback Consensus Tracking Control Of Non-affine Multi-agent Systems

Currently,with the rapid development and popularization of computer,network and communication technique,the system is presenting the characteristics of large,networked and distributed.The distributed cooperative control of multi-agent systems(MAS),as one of the effective way to solve the control problem of large-scale systems,has been attracting much attention and consensus control problem as the foundation and core of distributed cooperative control is also becoming a hot topic.In current,the distributed consensus control of affine nonlinear MAS are considered in most existing results.However,in some applications,such as wind turbine systems and the process of chemical reaction,the systems are modeled as non-affine form,in which the control input and system states are coupled together,which make controller design approach of affine system cannot be directly extended,so it's meaningful and changeable to study the consensus control problem of non-affine MAS.In this paper,the distributed tracking consensus control of non-affine nonlinear MAS are considered and two different control strategies are proposed such that the different system performance are guaranteed.The main results of this paper are as follows:(1)Regarding to the finite-time consensus tracking control for a class of non-affine MAS with only the system output available,a distributed output-feedback finite-time tracking control strategy is proposed.Firstly,by introducing a first-order low-pass filter and newly defined systems states,the non-affine MAS are transformed into affine one.Then,a finite-time observer is designed to estimate the states of the new derived affine system.By applying the backstepping method in conjection with fraction-order based DSC approach and RBF neural network,the output feedback based finite-time distributed tracking control laws are proposed and it proved that the consensus tracking errors converge to an arbitrarily small bounds in finite time.Finally,simulations are conducted to verify the effectiveness of the developed method.(2)Regarding to the distributed tracking control problem for a class of non-affine MAS with full state constraints,a distributed tracking control law is developed such that the full state constraints are not violated.Firstly,by applying a nonlinear mapping technique,the non-affine MAS with full state constraints are transformed into a novel one without full state constraints,which makes the controller easily.The Butterworth low-pass filter is introduced to avoid t algebraic loop,and then by combining with the backstepping method,DSC and RBF neural network,distributed tracking control law and adaptive law are developed.Finally,simulation results are provided to demonstrate the effectiveness of the proposed method.