UGV’s Vehicle Positioning Technical Research Based On Apriltags

Since the world’s first mobile robot Shakey came out in 1956,mobile robot technology has made great progress in many fields along with the development of related technologies,playing an increasingly important role in various fields.As a typical mobile robot,UGV(Unmanned Ground Vehicle)has attracted more and more attention.One of the key technologies in the field of UGV is the location problem.In contrast,the camera sensor based visual positioning method is widely used in the positioning research of UGV due to its advantages of low cost.Among them,Apriltags algorithm has been widely concerned in recent years due to both tracking and positioning functions.This paper mainly aims at the problem of using Apriltags tags to locate UGV,and makes an in-depth study of the correlation technology,and puts forward an improvement method to solve the problem of low accuracy of Apriltags tags positioning from two aspects.This paper first analyzes the Apriltags tag location algorithm,and in the experiment,it is found that when it has the long-distance measurement,the positioning accuracy is not high and the azimuth measurement results has deviations.Therefore,this paper proposes a nine-palace multi-tag system,which can improve the measurement accuracy of spatial pose by arranging multiple labels regularly in a 3*3 nine-palace grid.This method solves the problem of serious deviation of azimuth measurement results well.In order to solve the problem of poor accuracy of tag positioning at a long distance,a relative positioning system based on repeater tag is proposed.By setting repeater tags between the camera and the target tag,the long distance is divided into several short distances,so as to improve the measurement accuracy of the tag position.The above two studies mainly focus on the location of tags in the static state.In order to improve the accuracy of continuous tag positioning in the state of motion,this paper studies the adaptive Kalman filter in to explore its effect of improving the measurement accuracy of continuous moving tags with the above two tag improvement schemes.Finally,this paper combines the above three parts to build a programmable UGV with Raspberry PI as the core.The improved Apriltags positioning tags are loaded on the vehicle,and the UGV positioning technology based on Apriltags positioning tags and Kalman filter is studied and verified.