Coordinate Positioning And Formation Control For Multi Robot In Unknown Environment

As the main enabling equipment of intelligent manufacturing,robot has attracted extensive research and attention in the scientific and industrial fields,and the related technologies have developed rapidly and become more and more mature.With the improvement of the complexity of the application environment,a single robot sometimes can not complete the needs of complex tasks,so the coordinated control technology of multiple robots has gradually become a hot research field.In the process of robot coordinated control,the position and attitude between multirobot as the basic input is particularly important.In this paper,two wheel differential mobile robot is used as the carrier to study the coordinated positioning and formation control among multi-robot.The main aspects are as follows:(1)In order to complete the research of the problem,a reliable wheeled robot experimental platform is built,and a set of feasible experimental scheme is designed on the robot platform.It mainly includes: the construction of robot software and hardware,the selection and installation of sensors,and the design of bottom control system.(2)In order to coordinate the positioning of multiple robots and improve the positioning accuracy,this paper establishes the robot motion model on the experimental platform,which can calculate the horizontal and vertical distance of robot walking.Combined with the panoramic vision system on the top of the robot and the network data sharing system between robots,the coordinate axes between robots are solved and unified by gyroscope,and finally the robot observation model is established.Finally,the extended Kalman filter(EKF)algorithm is used to fuse the system data to obtain a more accurate distance between robots.The experimental data show that the positioning error is controlled within 6 cm when the calculated distance between robots is within 4 m.Compared with the positioning error of dead reckoning and panoramic vision system is more than 15 cm within 4 m,the positioning accuracy has been significantly improved.(3)On the premise of obtaining the relatively high-precision distance and angle between robots,the robot formation control is studied.One is the leader-follower formation mode based on the virtual structure.This formation mode belongs to centralized control and adopts a point stable following mode.When the distance between each other is far and the speed is 10 cm / s,the formation is formed at 25 s and the formation keeping accuracy error is controlled within 5 cm.The other is the swarm formation algorithm based on the swarm algorithm pilot follower.The algorithm assumes that at least one robot in the multi-robot system can track the virtual pilot followers.Without centralized control,through individual sensing around and making corresponding response behavior,the multi-robot system presents a certain coordinated formation mode as a whole.When the speed is 10 cm / s,the formation is formed in 15 s,and the accuracy error is also controlled within 5 cm.The feasibility of the above two algorithms has been verified on the experimental platform.(4)This paper summarizes and prospects the coordinated positioning and formation control of multi-robot system.