Research On Fully Distributed Cooperative Control Of Multi Agent System With Input Saturation

In recent years,with the rapid development of industrial technology,the cooperative control of multi-agent systems has been widely applied.The consensus problem of multi-agent systems is the most basic research problem in cooperative control,which has been widely concerned by researchers.Consensus control means that the state or output of agents can be achieved consensus through the design of controller and information exchange between agents.However,when designing the controller of agents,we often need to rely on some global information which is difficult to obtain,which is not conducive to practical application.In addition,the agents system whose control input is constrained by physical system often has saturation phenomenon,and the large amount of information exchange between agents will cause unnecessary consumption of communication resources.Therefore,it is of great theoretical and practical significance to study the consensus of multi-agent systems with input saturation based on event-triggering mechanism.Firstly,the problem of event-triggering consensus control for a class of uncertain multi-agent systems with input saturation is proposed.It is proved that the multi-agent systems can achieve semi-global consensus under the control strategy.The proposed low-gain feedback control strategy does not depend on the global information of the network topology and the size of the network.Then,the proposed control law is extended to uncertain multi-agent systems with input saturation based on dynamic event-triggering mechanism,thus eliminating the requirement of continuous communication.Under the event-triggering consensus control law,it is proved that each agent can strictly exclude Zeno triggering behavior,thus ensuring the physical enforceability of the proposed event-triggering mechanism.Then,aiming at the cluster output regulation problem of a class of heterogeneous multi-agent systems with input saturation,a fully distributed cluster consensus control strategy based on adaptive control and event-triggering mechanism is proposed,in which the adaptive control strategy is used to avoid the use of some global information(such as the minimum non-zero eigenvalue of Laplacian matrix related to communication topology,the number of multi-agent and so on).The dynamic event-triggering mechanism can further reduce the use of communication resources,and the influence of saturation nonlinearity on the system can also be eliminated by low-gain technology.Based on Lyapunov analysis and output regulation theory,it is proved that the system is stable and Zeno triggering behavior can be eliminated under dynamic event-triggering control law.Finally,through MATLAB simulation,the effectiveness of the proposed low-gain feedback control strategy based on event-triggering mechanism and the fully distributed cluster output regulation method is verified.