Research On Consensus Of Multi-agent Network Under Uncertainties

The coordinated control of multi-agent networks has been widely applied in the fields of unmanned aerospace,satellite formation,information fusion of sensor networks,the direction of movement of fish or birds,and distributed sensor filtering value,which has become one of the forefront research issues that scholars have paid attention to.The consensus problem of networks is the most essential issue in coordinated control and it plays a key role in the field of systems and control theory.Due to the influence and constraints of many uncertain factors,such as the time delay of information transmission between agents,the difference between the established system model and its actual structure,the interference caused by noise environment,etc.,it is meaningful to further explore the consensus of multi-agent networks.On the basis of predecessors' work,this paper researches the consensus behavior of multi-agent networks affected by uncertainties.The main contents are as follows:(1)The mean square average generalized consensus of multi-agent networks under fixed directed topology is analyzed,where all agents are affected by stochastic disturbances.Distributed protocol depending on delayed time information from neighbors is designed.By constructing Lyapunov functional,and sufficient conditions for multi-agent networks to reach mean square average generalized consensus are given by means of linear matrix inequalities.(2)The mean square average consensus problem for multi-agent networks with directed topology being strongly connected is investigated.Since agents are inevitably affected by noise during the process of transmitting information,the delay-based control scheme is considered to improve the control performance of the system.Based on Lyapunov stability analysis method,some sufficient conditions for the mean square average consensus of the multi-agent systems are given.(3)The consensus of linear multi-agent systems model with randomly occurring nonlinearities and uncertainties under distributed delay control is studied.Based on the distributed delay control strategy,which makes full use of the agents' delay information,the consensus analysis in the noise environment is performed.This more fully interprets the complex dynamic behavior of the networks.Some sufficient conditions are given to achieve the consensus of multi-agent networks asymptotically with linear matrix inequalities.(4)The consensus problem for delayed multi-agent systems with randomly occurring nonlinearities and uncertainties is considered under event-triggered control.In order to effectively lower the number of communication between agents and reduce unnecessary waste of resources,an event-triggered control scheme is designed to improve the control efficiency of the networks.The Lyapunov stability analysis method is used to obtain sufficient conditions for all follower agents in multi-agent networks eventually reach the state of leader.