Distributed Formation Control Strategy Design And Simulation Verification Of Multi-agent System

In recent years,multi-agent systems(MASs)have received extensive attention from researchers in various fields and have developed into autonomous,social and reactive agent systems.The research results of multi-agent cooperative control are widely used in key fields such as UAV swarms,military reconnaissance,resource exploration and disaster relief.Formation control is one of the core problems of cooperative control of multi-agent,and researchers mainly focus on the control strategy of formation to carry out in-depth analysis and discussion.A reasonable formation control strategy not only enables agent formation to achieve efficient mission cooperation,but also improves the reliability,robustness and adaptability of the system,which promotes the rapid development of multi-agent system.Considering the importance of formation control,this dissertation studies the formation control strategies of various models and explores the application scenarios of different models,while verifying them by simulation with several examples.The specific research of this dissertation is as follows.(1)A distance-based formation control strategy is proposed in this dissertation for the rigid formation control problem of first-order and second-order multi-agent models.By adding corresponding formation control terms,the multi-agent accomplishes three formation missions,such as formation acquisition,formation maintenance and formation dynamic transformation,respectively.Considering the possible rotation and tilt problems during the formation motion,the dissertation provides suitable translational and angular velocities and selection rules to realize the encircling motion of the formation.Meanwhile,through matrix theory and algebraic graph theory,the sufficient conditions for the stable operation of the formation task are obtained by using the energy(including kinetic and potential energy)related Laypunov function,and the asymptotic convergence of the formation error of the multi-agent system is realized.The feasibility of the formation control algorithm is verified by two numerical simulation examples.In the simulation verification,the feasibility of the control strategy is verified by designing different dynamic spacing between the leader and the follower to accomplish a variety of formation tasks.(2)To address the situation where global information is not available in multi-agent systems,a relative position-based adaptive control protocol is proposed in this dissertation to study the time-varying formation tracking problem of linear multi-agent systems.The control protocol does not rely on known communication topologies and can achieve formation tracking when global information is unknown by using only the relative output information between neighboring agents.Through algebraic graph theory and Laypunov’s stability theorem,sufficient conditions for linear multi-agent systems to achieve formation control are derived,and formation tracking of multi-agent systems is realized.Finally,the feasibility of the control strategy is verified by designing different dynamic spacing between the leader and the follower to accomplish a variety of formation tasks.(3)Taking into account the necessity of nonlinear models in practical engineering applications,an adaptive control strategy based on output feedback is proposed in this dissertation.By introducing a nonlinear term on the basis of the linear model and using the onesided Lipschitz condition to constrain the nonlinear term,the bipartite formation control in the presence of external disturbances is achieved.The control strategy depends on the state information of neighboring agents,independent of the control input of the navigator,and avoids the use of Laplace matrix eigenvalues.Further,the stability of the system is analyzed using Laypunov’s theorem to derive sufficient conditions for the stability of the system,and the bipartite formation control of the nonlinear model with/without disturbance is realized.The feasibility of the strategy in the bipartite formation task is verified by simulation examples with/without disturbances.