Event-Driven Consensus And Distributed Formation Control Of Multi-agent System

The problem of consensus and formation control for multi-agent systems(MASs)has attracted more and more attention,which has become a hot research topic in recent years.Consensus problem and formation control are based on the communication and interaction between multi-agents in the network.In fact,due to the constraints of the system itself and the influence of network bandwidth,the consensus and formation control of multi-agents are very challenging.Based on this,this paper studies the event-driven consensus control and distributed formation control of multi-agent system.Considering the update mode of controller,we study the event-driven consensus of MAS.Considering the node deployment,formation maintenance and obstacle avoidance,we study the distributed formation control.Based on these,the main research contents of this thesis as follows:Firstly,aiming at the problem of excessive waste of agents' own resources and network energy consumption,the non-uniform event-driven control mechanism is introduced to study the consensus control of the first-order MASs with saturation constraints.In order to reduce redundant communication between agents,A distributed event-driven control strategy is proposed,a linear driving function is given and a consensus control protocol is designed,the stabilities of the MASs is analyzed by using S-procedure method and Lyapunov stability theories,and the consensus conditions are obtained.Furthermore,the saturated consensus control in the case of input time delay is considered,and the distributed consensus control strategy is proposed,which has positive significance to save the energy of agent itself and network consumption in communication.Secondly,event-driven consensus control for second-order linear MASs is further studied.For the case of without considering input saturation,the condition of system consensus is obtained by using linear matrix inequality(LMI)techniques under given driving function.Then,the consensus of MASs under input time delay constraints is analyzed.Subsequently,the case of input saturation and without time delay constraints is considered,the consensus control of saturated systems is studied by using Lyapunov stability theories,and the consensus conditions are given.For saturated systems with time delay,the improved sector condition and Lyapunov-Krasovskii functional are used to obtain system consensus by solving LMI.The initial domains of saturated time-delay system are estimated by solving the boundary of differential equations.Thirdly,aiming at the energy consumption of multi-agent network communication in MAS,first of all,the method of generating persisting formation in mobile sensor network is studied.On the basis of minimum rigid graph,an algorithm for generating the optimal rigid graph is given.With the connection relationship between the known vertices and the edges in the optimal rigid graph,the directional operation method for vertices with different degrees is given.By using the directional operation of several edges and the reverse operation technique of the path,the optimal rigid graph is transformed into the optimal persistent graph.The relevant theorems and the basis for generating the optimal rigid graph are given.On the basis of guaranteeing the optimal rigid formation,the two-ways communication between agents are turned into one-way,which plays an important role in realizing the formation topology optimization and saving communication energy consumption.Fourthly,aiming at formation generation of multi-agent systems,the problem of nonlinear sliding mode formation control under local Lipschitz conditions is studied.For the first-order system,the integral sliding mode controller is proposed to realize finite-time formation;for the second-order system,the non-singular terminal sliding mode controller is given to achieve asymptotically stable formation;furthermore,based on the boundary layer,the super-twisting control algorithm is given to realize the second-order finite-time formation,which effectively reduces the impact of chattering on the system.Finally,aiming at the problem of formation obstacle avoidance and target coverage of mobile sensors,the strategies of obstacle avoidance control of sensors and full coverage of target area are studied.For the problem of sensor obstacle avoidance,the potential field function is introduced and the switching control input strategy is designed by combining sensor mobility techniques with composite stream function method,which according to 1-coverage topology structure to generate target formation while achieving obstacle avoidance control.For the problem of full coverage of the target area,the full coverage of the target region is realized by combining the full coverage strategy with the sensor rotation characteristics and the reasonable direction angle of the sensor,which has a positive impact on saving sensor energy and reducing network energy consumption.