Research And Design Of The Multi-agent Coordination Control Experiment System

The multi-agent system is the research focus of the cross fields of Artificialintelligence and collaborative control. The multi-agent research focuses on the coherencecontrol, cluster control and so on. So far, the majority of the research has focused on thetheoretical level, but the actual experiment and application systems are very rare. In thispaper, we design a multi-agent coordination control experimental system for algorithmsimulation.Firstly, the overall design of the multi-agent coordination control experimentalsystem is introduced. The experimental system includes four modules that are motioncontrol, wireless communication, and positioning and strategy control.Secondly, we designed the agent. After analyzing motion model of the actual agent,we designed the motor drive circuit and control circuit of the agent. We used a variationintegral PID control algorithm, achieved the Follow-up control between the integralcoefficient and the system, and improved the control accuracy of the system. Meanwhile,in order to meet the requirements of the agent information exchange between individuals,we designed wireless module for communication between the agent and the frameagreement for each agent.Then, we further study the positioning of the visual based on the experimental system,and propose two positioning methods, global positioning and visual self-positioning.Visual global positioning based on the RGB model, using a threshold method to identifyagent, we derived the mathematical relationship between the camera position andorientation coordinates, to ensure the real-time positioning accurately, make the systemmore applicability and real-time. Visual self-positioning used the HSL model instead ofthe RGB model, exclude the interference of positioning accuracy generated by thebrightness changes due to changes in light intensity caused by the identification point, andimprove the positioning accuracy of the agent, Meanwhile, alleviating the pressure of theagent number increased to a global positioning computer calculation.Finally, through debugging the experimental system, the experimental results are given. According to the non-linear characteristics of the system, we put forward the energydensity model. We designed four experiments that are the agent to fixed point, the agentfollow agent, the multi-hop communications and the multi-agent surround. Theexperiments show that the agent is controllability, the communication system has highstability, and the experimental system of multi-agent has coordination control ability, andlays the foundation for the simulation of multi-agent coordination control.