Position estimation using multiple low-cost GPS receivers for outdoor mobile robots

Abstract

Recently, the application of navigation system to electric devices such as cars, airplanes, and cellphones has become common. This common usage increases the demand for GPS, and therefore, GPS accuracy has become important. However, applying an external correction device or using a high-accuracy GPS receiver is very expensive; thus, practical applications of robotic systems using this technology are limited. Therefore, in this paper, we propose a method of using multiple low-cost GPS receivers in position estimation to enhance accuracy and decrease system buildup cost. The position estimation is based on the following steps: 1) sensor modeling of each GPS receiver, and 2) position estimation using extended Kalman filter based on the model of each sensor. The result of this study showed enhanced accuracy of the position estimation and the correlation between the number of GPS receivers used and accuracy of the position estimation. The position-estimation accuracy can be enhanced with low system buildup cost, and therefore, the proposed method can contribute to the field of robot control in outdoor environment.