Research On Adaptive Fixed-time Consensus Control Of Nonlinear Multi-agent System

In recent years,the fixed-time consensus of multi-agent systems has attracted the attention of many researchers.Compared with asymptotic consensus,fixed-time consensus has faster convergence speed,higher control accuracy,stronger anti-interference ability and broader development prospects.In addition,there are often problems such as input dead zone or actuator failure in the actual system,and due to the influence of the working environment,output constraint or full-state constraint on the system are required.Therefore,in view of the above situation,how to realize the design and analysis of multi-agent system consensus protocol has important theoretical significance and application value.Some results have been achieved,but there are fewer studies on fixed-time consensus of multimechanical systems and even more general high-order non-strict feedback nonlinear multiagent systems.Based on the fixed-time consensus theory,combined with power integrator technique and neural network,this paper studies the occurrence of actuator faults,input,output and full-state constraints in multi-agent systems,and proposes a series of new fixedtime consensus protocol design schemes.The main work of this article is as follows:(1)For multi-mechanical systems with unknown nonlinear characteristics and disturbances,a fixed-time consensus protocol based on adaptive neural network is designed.Firstly,considering the multi-mechanical system model with unknown nonlinear function terms and uncertain disturbances,the neural network is used to approximate the unknown nonlinear terms in the model,and a fixed-time consensus protocol is given based on the recursive design idea.Secondly,the Lyapunov stability theory is used to prove that the position and velocity errors of any two mechanical systems can converge to a small neighborhood of zero in a fixed time.Finally,numerical examples are used to verify the effectiveness of the designed protocol.(2)Considering the multi-mechanical system with actuator fault and output constraint,a fixed-time fault-tolerant consensus protocol is proposed.Combined with graph theory,nonlinear mapping method,neural network and power integrator technique,an adaptive practical fixed-time consensus protocol is designed when the actuator partial failure and bias fault exist at the same time,and numerical simulation verifies that the protocol can make the two states of the multi-mechanical system achieve consistency within a fixed time and the position state always within the constraint.(3)The practical fixed-time consensus problem of high-order nonlinear multi-agent systems with both input dead zone and full-state constraints is studied.Considering that the multi-agent system has a non-strict feedback structure and is limited by the asymmetric time-varying full-state constraint,the new nonlinear mapping function is used to transform the constrained system into an equivalent unconstrained system,combined with power integrator technique and neural network,a recursive design scheme of adaptive practical fixed-time consensus protocol is given,and the fixed-time consensus analysis of multiagent system is carried out by using Lyapunov stability theory.Finally,numerical simulation is carried out,and the simulation results show that the follower state can be tracked to the leader within a fixed time and all the state signals never violate the constraints.