On Reaching Group Consensus Of Multi-agent Networks With Multiple Time Delays

In recent years,due to the widely applications in unmanned aerial vehicle formation control,sensor network congestion control,distributed energy Internet control and other fields,the consensus study of dynamic multi-agent network exploration has attracted great attentions of researchers at home and abroad.Usually,the consensus means that without any control of global communications,all agents in a network through limited local interaction adjust their behavior states,eventually,all agents’ behavior states have to reach consensus.Meanwhile,consensus states of agents can be denoted by position,altitude,temperature,speed and other variables.As a fundamental issue in distributed cooperative control of complex multi-agent networks,the exploration of consensus topic is to better understand the inherent evolution rules of complex systems and to analyze the characteristics of dynamic complex systems with external environment.Therefore,the research work has great impetus to the development of complex systems.However,with the changes of external environment,cooperative tasks and even time,the consensus states of multi-agent networks will be changed.Apart from that,in multi-agent networks,when different target tasks are completed harmoniously by multiple agents,it can also present distinct consensus results.Generally,the phenomenon can be described by the group consensus of multi-agent networks.At the same time,along with the continuous expansion of network structure complexity and nodes’ number,the difficulty and complexity of consensus control also will increase.Therefore,it is more urgent for us to design a suitable and effective group control protocol to study the group consensus of multi-agent networks.This thesis mainly investigates the group consensus for multi-agent networks under the influence of multiple time delays.The systems include first-order,second-order continuous and discrete as well.The main research work is listed as follows:For networks with the topology which is a connected bipartite graph or contains a globally reachable node,some constraint conditions are proposed analytically,which can guarantee the continuous and discrete multi-agent networks reaching group consensus.From the theoretical analysis and experiments results,it is found that the achievement of group consensus is independent of communication delays,but it can be determined by input delays and coupling weight between nodes.Different from the first-ordercontinuous systems,the realization about group consensus of second-order continuous systems is also affected by systems’ coupling strengths.Compared with second-order continuous systems,the group consensus of second-order discrete systems is also determined by input delays and systems’ coupling strengths.Meanwhile,the reaching about group consensus of second-order discrete systems also needs to meet the algebraic relations of systems’ coupling strengths.Simultaneously,it is also interesting to find that communication time delays between agents can interfere with the consensus rate of discrete systems.