Research On Multi-agent Systems Consensus Via Impulsive Control

In recent years,with widespread applications of intelligent devices and rapid developments of communication technology,networked multi-agent systems have received increasing attention.Consensus is an important issue in the framework of multi-agent systems,and the impulsive control has advantages and features such as simple controller structure,strong robustness,reducing information redundancy and accelerating system response time.However,in academic research on impulsive consensus,further exploration is needed for consensus analysis of different dynamic models under non-ideal conditions with fixed impulses,asynchronous impulses and time-varying window impulses.In addition,there has been limited research on consensus of open systems.The problem of inconsistent state vectors caused by openness has yet to be resolved which prompts us to undertake further discussion.This article mainly focuses on following research content:(1)The impulsive bipartite consensus problem of second-order multi-agent systems with external disturbances was studied.For a class of second-order nonlinear multi-agent systems subject to bounded or stochastic disturbances,a control method combining data sampling with impulses was proposed.Meanwhile,the communication data were subject to saturation conditions,and the saturation function term in the control protocol was handled through convex hull theory.Based on It(?) stochastic differential equation theory and Lyapunov stability theory,sufficient conditions for leader-following bipartite bounded consensus were derived,where the error upper bound depends on the external disturbance norm.Furthermore,the system without external disturbances was considered,by using average impulsive gain to further reduce conservatism of sufficient conditions mentioned above.The feasible range of impulsive signal intensity was expanded,and it was verified that disturbances had a negative impact on system performance.(2)The asynchronous impulsive consensus problem of heterogeneous second-order multi-agent systems on time scales was studied.For a class of second-order heterogeneous multi-agent systems,an asynchronous impulsive control method was proposed to set impulsive time series individually for each node.Furthermore,with the help of given variables,a novel analysis method was proposed,which can effectively fuse several impulsive time series into one time series.In addition,for leader-following asynchronous impulsive consensus,a scheme was proposed to sort error vectors according to the norm.Nodes with larger error norms were selected for implementing impulsive control.The quantized numerical relationship between the number of pinning nodes and the number of global nodes at different impulsive instants was obtained.Furthermore,based on time scales theory and Lyapunov stability theory,sufficient conditions of static leader-following consensus and leaderless consensus on time scales were obtained.(3)The window impulsive consensus problem of time-varying multi-agent systems on time scales was studied.For a class of time-varying multi-agent systems subject to deceptive attacks on time scales,a method combining data sampling with time-varying window impulses was proposed.The time-varying impulsive window relieved the constraint of setting fixed impulsive instants in advance.For deception attacks on transmission channels,a random variable obeying Bernoulli distribution was proposed to describe the attack occurrence and its probability.Based on time scales theory and Lyapunov stability theory,sufficient conditions for leader-following consensus and dynamic average consensus dependent on the average parameter related to impulsive gain and time-varying window were obtained,which not only reduces the input frequency requirement of impulsive instant,but also reduces the constraint range of the impulsive gain.(4)The impulsive dynamic average consensus problem of time-varying open multiagent systems on time scales was studied.For a class of time-varying open multi-agent systems on time scales,a control method combining data sampling and impulses was proposed,and the open instant was set as the sampling instant in advance.Firstly,nodes were divided into three categories,and the concept of open state vector was proposed to solve the problem of vector change before and after the opening instant.Based on time scales theory and Lyapunov stability theory,sufficient conditions for dynamic weighted average consensus were obtained.In addition,when the system involves unmeasurable information,the impulsive adjustment term was constructed by output vectors of two systems.The impulsive observer was introduced to reconstruct the original system information and ensure it can reach dynamic weighted average consensus.(5)The prescribed-time impulsive consensus problem of time-varying open multiagent systems with delays on time scales was studied.For a class of time-varying open multi-agent systems with delays on time scales,a control method combining a novel piecewise state feedback control protocol combined with impulses was proposed.At each open instant,an impulsive signal which dependent of the leader was introduced to solve the problem of too large controller norm.By extending Halanay-like inequality on time scales,the difficulty of theoretical analysis caused by time delays was solved.Based on time scales theory and Lyapunov stability theory,sufficient conditions for prescribed-time consensus were obtained.In addition,when the absolute information of each agent cannot be obtained completely,the observer system was designed to reconstruct the information of the original system and ensure it can reach consensus within the pre-set time.