Consensus Of Multi-Agent Systems Under Incomplete Information

The study of the multi-agent systems is a hot subject in the field of automatic control.Compared with traditional control system,the multi-agent system is higher efficiency,wider application range and lower cost.Meanwhile,robustness,flexibility and other characteristics are well presented through network sharing resources,Therefore,the research on the network multi-agent systems is highly valued by workers.As the foundation of distributed collaborative control of multi-agent systems,the study of consensus is of great significance to the field of collaborative control.The multi-agent systems transmit information through the network to achieve resource sharing among various agents.However,due to the double-sided nature of the network,the multi-agent systems will also have a series of inevitable challenges when communicating.For example,it is necessary to overcome problems caused by the various limitations of the network itself during the transmission of information,such as communication delays,packet loss and so on.These conditions will reduce the performance index and stability of the system,which will affect the analysis and design of the controller.Therefore,the research on the consensus of multi-agent systems under incomplete information is of great scientific significance.Firstly,the problem of large system delay caused by large data delay and loss is aimed in this dissertation,discrete time multi-agent system model is adopted,Predictive control methods are used,the effect of delay and packet loss on the agent system is actively compensated.Meanwhile,graph theory and matrix theory are used to calculate the consensus controller of multi-agent systems under incomplete information.In addition,the conditions for the stability of networked multi-agent systems are discussed in this paper.Secondly,for the case where the number of delays and packet losses is less than a certain range,a continuous-time multi-agent system model based on a periodic sampling mechanism is used in this paper.Meanwhile,considering the delay caused by different situations,the conditions for the multi-agent systems to achieve uniform sampling and the maximum delay and packet loss rate that the system can bear are obtained by the switched system theory and Lyapunov function method.Lastly,the validity of the results are verified by numerical simulation.