Distributed Tracking Control Of Strict-feedback Non-affine Multi-agent Systems

In recent years,the distributed coordination control of multi-agent systems(MAS)is a hot research topic,which has gained much attention due to its broad application backgrounds in many disciplines and fields,such as computer sciences and biology.The consensus control problem is the most classic problem in MAS coordination,and it can be roughly divided into two research directions: the consensus problem without leader,the other one is the tracking control problem with the leader-follower.The leader-follower tracking control problem has made a lot of achievements,but most researches are regarded to the affine MAS.However,a class of practical systems such as chemical reaction process,wind energy conversion systems,Duffing-Holmes chaotic system,and air-breathing hypersonic vehicle are described in the non-affine form.So,it is necessary to study the non-affine MAS.In this paper,we study the distributed tracking control algorithm for non-affine MAS,and propose two different types of distributed tracking control strategy.The main contents are as follows:(1)Regarding to the leader-follower problem of the strict-feedback non-affine MAS with the states can be measured,we proposed a distributed state-feedback tracking control algorithm to realize the consensus tracking of the followers and the leader under directed topology.Firstly,By combining a low-pass filter and the state variable transformation,the non-affine systems are transformed into the affine form.Then,the full dimension state observer is designed to estimate the new unknown states.RBF neural network is used to approximate the unknown nonlinear function in the design,and through the Backstepping method and dynamic surface(DSC)technology to design the controller.Finally,simulation results are provided to demonstrate the performance of the developed controller.(2)Regarding to the leader-follower distributed consensus tracking control problem of the strict-feedback non-affine MAS with partial unmeasured states,we proposed a distributed output-feedback tracking control algorithm to realize the consensus tracking of the followers and the leader with Prescribed performance under directed topology.Compared with the existing results,the restrictive assumptions on the partial derivative of non-affine functions are removed.Since the control input is implicit for each non-affine agent,we introduce an auxiliary first-order dynamics to circumvent the difficulty in control protocol design and avoid the algebraic loopproblem in the design process.A fully distributed input-driven observer is applied to reconstruct state information of each agent,which reaches the estimation of states.Finally,Based on the dynamic surface control(DSC)technique and RBF neural network approximators,a consensus control protocol with prescribed performance bounds(PPB)is derived.The simulation results are provided to demonstrate the effectiveness of the proposed method.