Research Of Localization Technology On Indoor Robot Based On EKF

Autono mous mobile robotics research is a significant branch of robotresearch area. The primary prob lem of designing autonomous robots is robotnavigatio n proble m. In order to solve the robot navigatio n proble m, theproble m of positioning of the robot must be solved. A localization syste m mustbe able to determine the re lative position between the robot and the startingpoint. And la y the foundatio n for robot navigation. The ma in contents of thispaper is to study mobile robot localization problem.The purpose of this paper is to design a localization syste m of indoorrobot based on embedded operating syste m. The workflow o f this s ystem is asfo llows: F irst, give the robot a training path, and the robot records the trainingpaths, then the robot move in accordance with the path as trained. In addition,in order to ensure the independence of the robot can work continuous ly, thisresearch designed a robot automatic charging system, when the battery powerof the robot fa lls below the thresho ld, the robot will find the char ging baseautomatically.The motion of the robot could be controlled by the odometer and compasswhen the robot moves in the training path. In this way, however, the robot ma ydeviate the training path as time grew, since the odometer brings inaccumulated error. In this paper, the robot comb ined the information o fodometer and signpost to eliminate the accumulated error. In this way, therobot faces the problem o f data fus ion. This paper uses EKF(Extended Ka lmanFilter)algorithm to solve this problem.Specific work of this paper is as follows: First, This paper made anoverview o f the background and development of autono mous robots androbotic positioning techno logy, the n, elaborated the working princ iple anddesign process of setting up the wheeled robot b ased on embedded syste mplatform and the robot modules required for positioning in detail.Fina lly，This paper has studied the Kalman filter algorithm and extendedKalman filter algorithm in robot loca lization and e laborated detailed process of EKF(Extended Kalman Filter) robot localizatio n and made simulations ofrobot localization us ing EKF on. The result of simulation shows theeffectiveness of extended Kalman filter algorithm in robot localization.