EKF Based GPS/Odometer Data Fusion For Precise Robot Localization In Outdoor Environment

Abstract

This paper describes a method of data fusion for outdoor autonomous robot equipped with onboard GPS receiver and Odometer using Extended Kaman filter. Both GPS and Odometer having their own advantage and disadvantage for accurate position information of Robot in unknown/outdoor environment. Described Data fusion technique fuse position input of both Odometer and GPS to generate most accurate position estimation with less uncertainty (standard deviation) by using Extended Kalman Filter. Once precise position of Robot obtained in unknown environment then other control actions can be applied on Robot to put it on the right track/Path. Simulation of fusion technique and all results has been done in MATLAB with proper units and values.