Research On The Leader-following Consensus Of Multi-agent System Based On Sampled Output Measurements

With the rapid development of computer technology,the sampling control receives more and more attention.Many achievements have been made in the research of sampled-data control systems over the decades,but more studies are based on linear systems.Research on nonlinear systems are relatively few,which need futher study.There are always some problems in the process of network sampling and transmission,such as packet loss,transmission delay and quantization error,and so on.In order to minimize the impact of these problems and improve the universality of the research results,it is necessary to study the sampling control systems.Based on the Laypunov stability theory and sampled-data system theorey,we discussed the consistency problem of multi-agent system and the stability of event triggered network control system,which are based on the sample output and communication lag,respectively.The main contents of this paper are summarized as follows:First,the consistency of upper triangular leader following multi-agent system based on the communication lag and sampled output is considered.Different from previous research work,this paper releases the restriction condition that the transmission lag is less than the sampling period.The transmission lag can be larger than the sampling period.A sampled observer is constructed based on the sampled output,then the output feedback controllers are proposed.Lyapunov functional is constructed based on the Lyapunov stability theory,and sufficient conditions are given to ensure the considered multi-agent system achieves global and exponential consensus.Similarly,observer-based dynamic output feedback controllers are propodsed for the lower triangular multi-agent system,and sufficient conditions are given to ensure the system achieves leader-following consensus.Second,the consistency of upper triangular leader following multi-agent system is futher studied.In this part,two kinds of transmission lags during data transmission are considered,i.e.,the delay from the system to the output feedback controllers and the delay from the controllers to the ZOH(zero-order holder).Based on the sampled output,the sampled observer and output feedback controllers are constructed.The interval decomposition method is used to decompose the interval,such that the controller and the multi-agent system update their states on the same interval.Finally,sufficient conditions are given to ensure the considered multi-agent system achieves leader following consensus.Similarly,observer-based dynamic output feedback controllers are proposed for the lower triangular multi-agent system,and sufficient conditions are given to ensure the system achieves leader-following consensus.At last,we summarize the results of the dissertation and discusse the research direction that can be improved in the future.