The Research On Cooperative Formation Control Of Multi-agent Systems

With the development of communication technology and computer technology,robots,unmanned vehicles,unmanned surface vehicles and other intelligent agents have been widely developed and applied in military and civil fields.The working ability of one agent is usually limited.One agent is unable to meet the needs of complex tasks such as large object capture and large-range airspace investigation.However,these tasks can be completed effectively by using multiple agents to work together.Multi-agent systems demonstrate the ability of higher task execution efficiency and better fault-tolerant ability.Therefore,cooperative control problems of multi-agent systems have gradually become a hot topic.As one of the important research problems of cooperative control problems of multi-agent systems,formation control problems have received more and more attention.Traditional formation control methods usually neglect the settle time of formation.However,in practical applications,systems are usually required to complete tasks in a finite time.Compared with traditional formation control methods,finite-time formation control methods can make systems form a desired formation in a finite time,so they are more suitable for practical application systems.However,finite-time formation control methods depend on initial states of systems.Thus,they cannot be widely used in practical applications.Therefore,fixed-time formation control methods should be developed.Fixed-time formation control methods are independent on initial states of systems so that they are more practical.Taking the importance of formation control problems into consideration,the formation control of discrete multi-agent systems,the formation control of multi-agent systems in fixed time,the core sensing data processing and the cooperative formation experiment of unmanned surface vehicle systems are discussed.The main research contents of this thesis are given as follows.(1)For a multi-agent formation system described by a discrete-time dynamic model,an iterative learning based collaborative control method for the multi-agent formation system is proposed in this thesis.Firstly,considering the external environmental disturbance,the discrete-time model of the multi-agent formation system is established.Then,by designing a leader-follow communication topology,the cooperative formation control problem of the multi-agent system is transformed into a tracking control problem.A unified formation control protocol based on an iterative learning algorithm is designed for every agent.The convergence of the algorithm is proved by using norm theory.Finally,the effectiveness and superiority of the proposed method are verified by a simulation example.(2)For a single-integrator multi-agent system,a multi-agent fixed-time formation control method based on consensus theory is proposed to achieve the aim of fixed-time formation control.A formation problem of the multi-agent system is transformed into a consensus problem.Then,a tree-type transformation is used to transform the formation problem into a origin stability problem of another reduced-order system.A uniform fixed-time formation controller is constructed for each agent.The proposed fixed-time formation controller ensures that the actual formation of each agent and the expected formation is consistent at a fixed time,and the formation time is independent on initial states of systems.The validity of the proposed method is demonstrated by numerical simulation.(3)Taking unmanned surface vehicle systems into consideration,an experimental system of multi-agent cooperative formation is constructed.Then,data acquisition and processing methods of unmanned surface vehicle systems are proposed.Then,the formation control experiment is carried out.The data obtained by sensing equipment of unmanned surface vehicle systems are processed by using Kalman filter.Then,a corresponding data processing algorithm is developed to extract the actual required position and attitude data.After that the information such as the position and attitude of unmanned surface vehicle systems is smoothed.Then,the processed location and attitude data are applied to an IPC(industrial personal computer),and a suitable control is used to generate the corresponding control instructions,so that unmanned surface vehicle systems can form the desired formation.Finally,the validity and correctness of the proposed method are verified by experimental results.The experimental results show that unmanned surface vehicle experimental systems constructed in this thesis is effective.