Reserach Of The Formation Of Multi-agent Systems Based On UWB Positioning System In Complex Environment

Formation is an important cooperative behavior of multiagents.The study of multiagent formation control in complex environments has important theoretical significance and application value.In order to obtain a formation control method for indoor complex environments,this paper designs and optimizes a formation control method based on relative position information uses UWB positioning system for indoor positioning of mobile robots,and built an indoor formation control system platform for experimental verification.The main contents of the paper are as follows:1)A formation control method based on relative position information and virtual navigator is designed.The difference between the expected formation data and the actual positioning data is used to generate a virtual navigator to guide the robot's movement behavior to complete the generation of formation.The controller proposed in this paper has a simple structure,a small amount of calculation,and introduces an algorithm module for collision avoidance based on artificial potential field.It shows good formation performance and collision avoidance ability in the formation control platform.2)A method of formation maintenance based on the intermediate reference formation is designed to avoid the formation lost of the multi-robot system during formation rotation,zooming,translation and other movement processes.By simulating the process of maintaining the desired formation,the formation of the multi-robot system is smoother and more stable.Further by designing the fitness function of the formation process,the genetic algorithm is used to optimize the control parameters of the formation process,which effectively avoids the shock phenomenon in the formation process and further improves the performance of the collision avoidance module.3)Design and build a multi-robot positioning and formation control physical platform,and verify the rationality,effectiveness and reliability of the formation algorithm design through experiments.