Multi-robot Coordination Positioning Based On Panoramic Vision

With the development of global science and technology,mobile robot technology has been rapid development.At present,mobile robot has been gradually applied to all walks of life,and plays a positive and important role.However,with the improvement of the application environment of mobile robot,the previous single robot can not meet the requirements of complex tasks,so multi-robot cooperation technology has become one of the hot spots of robotics research.Multi-robot cooperation is inseparable from the acquisition of the location information between robots,so this paper mainly studies the multi-robot coordinated positioning system based on panoramic vision,which aims to complete the identification and localization of multi-robot and improve the positioning accuracy between robots.In this paper,the reliability and stability of the robot target recognition and localization,adaptive token ring network communication,data fusion positioning and related methods are verified through practical mobile robot platform.The specific research is as follows:Firstly,according to the requirements of mobile robot flexible,cheap price,easy to control,this paper designed and built mobile robot platform,the specific design of robot program,including the overall hardware and software framework design,sensor selection and configuration,bottom motion control system design,adaptive token ring communication design.Secondly,on the robot panoramic vision system,the specific process of visual processing.In the recognition process of Target Body,this paper designs a simple structure,replaceable robot Target Body,and gives a panoramic imaging recognition method based on polar,the method is simple and effective,and can quickly identify the feature points of multiple target robots.According to the feature points of the robot Target Body,an improved k-means algorithm is proposed,which can quickly and effectively get the coordinates and angles of other robots.Through the experimental calibration,the actual distance and angle of other robots relative to themselves can be obtained,and the real-time and stability of the ling network can be verified.Again,for the multi-robot coordinated positioning and positioning accuracy improvement,this paper established the robot bottom motion model based on the deadcalculation.According to the network information sharing and the multi-objective situation of the visual system,this paper uses the electronic compass coordinates to solve the matching,establishes the robot observation model,combines the robot motion model and observation model,and finally uses the extended Kalman algorithm to fuse each model data.Finally,this paper summarizes and prospects the multi-robot cooperative positioning system,and hopes to lay the foundation for the later research multi-robot cooperation through the scheme and algorithm proposed in this paper.