The Cooperative Control Protocol Design And Analysis Of Multi-agent System In Two Dimensional Plane

Multi-agent system is a hot research topic in the control science field. Because of its strong reliability, high efficiency and good expansibility, multi-agent system has been widely used in the practice.The consensus problem is the basic cooperative control problem for the multi-agent system.All agents in the system can reach the same state under the effective consensus protocol.Flocking and formation control are the most common consensus problems in the two-dimensional plane.In this thesis, local control method is used to solve the flocking problem. A local controller is designed to achieve the switching of the consistent balance state. The basic idea of the method is to develop a different control scheme by selecting different combination of agents, then to calculate the control parameters base on the original and new location information of the balance value, and finally to apply the local controller to the chosen agents so that the switching of consistent balance state be achieved. The distributed control method of similar formation is presented to solve the two-dimensional planar problem of formation. The designed distributed controller can realize a similar formation of the multi-agent system. Firstly, the formation control problem of multi-agent system is transformed into the stability problem of the multi-agent system. Then, the multi-agent system is decomposed into several subsystems, and the stability of each subsystem is analyzed. After that, the proportional controllers are designed to ensure the stability of the multi-agent system. The control parameter to make the entire system stable can be obtained from the intersection of the stability range of each subsystem. Finally, the simulation examples reveal the effectiveness of the designed controller.In order to realize the rapidness of the formation control, a distributed minimum-step algorithm in the two-dimensional formation is presented. Based on the matrix theory and the final-value theorem, a static distributed minimum-step algorithm and a dynamic distributedminimum-step algorithm are obtained, respectively. The advantage of the proposed algorithm lies in that each agent can predict the location of its next moment only according to its own former location information. Moreover, the location of the final formation can be calculated directly by using the static distributed minimum algorithm, which raises the speed of formation.