Distributed Consensus Control Of Multi-agent Systems Under Unmeasurable State And Constrained Communication Resources

In recent years,the consensus control of multi-agent systems has become a research highlight in the field of control due to its widespread application in power systems,intelligent transportation,spacecraft formation,multi-robotic arm collaboration,wireless sensor networks,and other fields.In order to achieve the goal of consensus control in multi-agent systems,information transmission and interaction between agents are carried out through the network.The limited communication resources such as network bandwidth and node energy in the network environment have brought problems and challenges to consensus control.In practical systems,due to the limitations of system structure and other factors,the full state information is difficult to measure directly or the measurement cost is high.Therefore,to design the output feedback controller using the output information to achieve the consensus control goal under the condition of an unmeasurable state has important theoretical and practical value.In addition,considering the operation process of a multi-agent system in the network environment,problems such as input saturation,actuator failures and network attacks bring further challenges to the realization of consensus control for multi-agent systems and even cause system instability.Therefore,considering the constraint factors such as limited communication resources,unmeasurable state information,network attack,input saturation,etc.comprehensively,the leader following consensus control problem of nonlinear multi-agent systems with unmeasurable state and limited communication resources is researched deeply in this dissertation.The main research work is as follows:Firstly,for the distributed finite-time consensus control problem of multi-agent systems with unmeasured state and actuator fault,the finite time observer is used to estimate the unmeasurable state and actuator fault simultaneously.Based on the supertwisting algorithm and improved integral sliding mode control,a fully distributed finite time consensus controller is designed to reduce the chattering and improve the flexibility and scalability of the system.By using the Lyapunov stability theory,it is proved that the consensus of multi-agent systems can be achieved in finite time.Secondly,for the distributed consensus control problem of multi-agent systems with unmeasurable states and limited communication resources,an adaptive sliding mode observer is designed to estimate the unmeasurable state.Based on the real-time state of itself and the triggered time state of its neighbor agents,a distributed leader-following consensus controller is designed to reduce the amount of information transmission between agents effectively.By Lyapunov stability theory,it is proved that the consensus of multi-agent systems can be achieved.Thirdly,for the multi-agent systems with the problem of limited communication resources and input saturation,based on an event-triggered mechanism and integral sliding mode control,a fully distributed finite-time consensus controller is designed to avoid continuous communication between agents.A second order auxiliary variable is constructed using saturation error signals to compensate for the adverse effects of input saturation on system performance.Through Lyapunov stability theory,it is proved that the consensus of multi-agent systems can be achieved in finite time.Finally,for the distributed consensus control problem of multi-agent systems with limited communication resources,the information of the leader can only be obtained by a group of followers,deception attack attack and Denial of service(DoS),an adaptive estimation of deception attack is designed.Then,using the estimated information of the deception attack,an event triggered distributed estimator is designed for each agent to estimate the state of the leader information,and the amount of information transmission between agents is reduced.For the changes of communication topology caused by DoS attacks,the average dwell time method is used to provide sufficient conditions to ensure the convergence of estimation errors,and the maximum tolerable duration of the DoS attack is obtained.Based on the error between the state of each agent and the estimated state of the leader,a leader-following distributed consensus controller is designed.By Lyapunov stability theory,it is proved that the consensus of multi-agent systems can be achieved.