Formation Control And Robustness Analysis Of Mobile Multi-agent Systems

In recent years, with the rapid development of computer and communication technology,network technology, sensor technology and artificial intelligence, coordination and control ofmulti-agent system has become one of the hotspots in the field of automatic control andcomputer applications, and gradually shows its interdisciplinary trend. One general definitionof formation control of multi-agent system is that: to keep the formation of multi-agentsystem when getting close to the target through distributed control. Formation control ofmulti-agent system is particularly important when coordinating multiple independentindividuals together to complete a complex task in the production process of practice.In this paper, the formation control and robustness of mobile multi-agent system areanalyzed. Firstly, distributed control protocols of formation system are adopted to realize theconsensus of the system through analysis of the motion of dynamic multi-agent system;besides, factors that can affect the stability are analyzed, and control protocols are provided toadapt to the uncertainties of the model. The threshold of the tolerance is offered, and as aresult, the formation control and anti-disturbance robustness of mobile multi-agent system arerealized.The following aspects are carried out on the basis of reviews of the correlated research inthis paper.1. Consensus of Multi-agent system with structure uncertainty and time-varying timedelays. In order to get the simple mathematical form, Kronecker product is used toabstract the expression of second order multi-agent system. To get the threshold oftime delays, reasonable Lyapunov function is constructed and the LMI approach isused, considering the symmetrical communication delays. To reduce the conservative,the idea of the free weight matrix is introduced in the argument of the mainconclusions. Further analysis is considered with the existence of structureuncertainties and the conclusions are expanded. Conclusions of consensus are givenwhen the topology is switching. Finally, simulation examples of the given systemmodel are provided.2. Convergence analysis of multi-agent system formation with a single virtual leader.The second-order multi-agent formation control system with time delays and a single virtual leader is considered, convergence analysis of multi-agent system formation isgiven through the analysis of different realizations of the formation control ruleswhich are proposed by Reynolds, and regulations for preventing collisions are alsoconsidered. The rules of obstacle avoidance are realized through the rational design ofdistributed control protocols, and a theoretical proof is provided through the design ofartificial potential functions. Simulations are given to prove the correctness ofconclusions and the switching topology is considered.3. Formation control of multi-agent system with a single virtual leader. According toprevious literatures, a new formation control model with a single virtual leader is putforward. There is not a direct link between the virtual leader and every agent in theformation system, only few of them have the information of the virtual leader whileothers get the information through their neighbors who know the information. Theupper bound of the delays is derived through LMI approach. Simulation examples aregiven to get the upper bound of the iteration time delays.4. Robustness analysis of the formation control system with a disturbance observer. Dueto inevitable outside interferences, the widespread presence of internal uncertaintyand modeling error, there is bound to be a variety of disturbances in the modelabstracted from the real world. A disturbance observer is constructed through theanalysis of disturbance of the system to make the control system more robust.Conditions for stability are acquired through LMI approach. Finally, simulationresults show the effectiveness of the method.