Research On Consensus Of Multi-agent Networks Under Event-triggered Communication

Distributed cooperative control of multi-agent networks has a very broad application prospect.For instance,in the field of industrial and military,there is a wide range of cooperative control of multi-agent networks such as cooperative control of unmanned aerial vehicle,estimation over wireless sensor networks,attitude control of airplane carrier and so on.Control theory and method of networked coordinated control systems are the core theory and technology of the above engineering applications.Among them,as a fundamental problem,the consensus problem of multi-agent networks has received attractive attention.Due to the constraints of network communication resources and the limitation of the computing power of the agent,conventional consensus control protocol which is based on continuous real-time information interaction can't be well guided for practice.Therefore,how to achieve the purpose of system control on the basis of sampled information has become one of the urgent problems to be solved in the field of consensus based control.The introduction of event-triggered communication mechanism provides a positive solution to the above issues.Based on this idea,our dissertation studies the event-triggered communication consensus of multi agent network.The main contributions and originality contained in this dissertation are as follows:(1)Under the conditions that each agent has the output delay and communication bandwidth of multi-agent networks is limited,we consider the reduced-order observerbased consensus problem of multi-agent networks with output time delay and event trigger strategy.First,a multi-step algorithm is presented to construct a reduced-order observer for each agent.Then,a novel push-based event-triggered control strategy,which based on the reduced-order observer and the relative outputs of neighboring agents is proposed.Under this control strategy,a sufficient condition for the consensus of multi-agent networks is obtained by using the integral inequality technique and matrix theory.Moreover,the estimation value of the output time delay is also obtained and the Zeno-behavior of triggering time sequences is excluded.Finally,two simulations about the multi-agent networks are provided to illustrate the correctness of theoretical results.(2)The event-triggered consensus problem of discrete time second-order multi-agent networks under the condition of limited communication bandwidth is concerned.First,a novel distributed discrete time event-triggered scheme is proposed.The controller is updated only at its own triggering time instants,which depends on the position and velocity.Then,by using bilinear transformation technique,the integral inequality technique and matrix theory,some sufficient conditions for the consensus of multi-agent networks are obtained.Different from existing works,our conditions reveal how the Laplacian matrix eigenvalues and scaling strengths affect the consensus of discrete time multi-agent networks.Moreover,the Zeno-like behavior of triggering time sequences is excluded.Finally,a numerical simulation about the multi-agent networks is provided to illustrate the correctness of theoretical results.