Multi-variable Trajectory Planning And Cooperation Control For Multi-agent Systems

Multi-agent technology has been widely used in industrial, military, economic andother fields. Multi-agent technology can simplify the modeling of complex systems, andanalyzes the performance of whole system. In this paper, the main content is themulti-variable control of multi-agent systems. The individual agent is the basiccomponents and the research base of multi-agent systems, while the key content is thecoordination and cooperation of individual agents.Firstly, for the trajectory planning control of individual agent, a design method ofmodel-free adaptive trajectory tracking controller is proposed based on the unfalsifiedcontrol method. For trajectory planning control law of individual agent, using theunfalsified control method based on data-driven, design a model-free adaptive trajectorytracking controller by the coordination control of line velocity and angle velocity. Thiscontroller can achieve tracking the desired trajectory rapidly and accurately, and avoid theobstacles.Secondly, a multi-variable flocking control law is proposed based on the receivedsignal strength indicator (RSSI). The RSSI is introduced to estimate the inter-distancebetween agents, to solve the problem that the relative distance between agents cannotaccurately obtained. A multi-variable integration parameter that contains the velocity,direction and RSSI of agents is defined, and a multi-variable controller is proposed basedon the parameter, to realize the flocking of multi-agent systems.Thirdly, for the requirement of some missions, the multi-agent systems need to keep acertain formation. Through the study of coordinate agreement and ID allocation, adistributed formation algorithm is proposed based on the local information exchange ofagents. This formation algorithm can form pattern rapidly, and maintain the stability of theformation.