Kinematic GPS positioning in a highway environment

Abstract

The global positioning system (GPS) is a satellite based navigation system which can give twenty-four hour three-dimensional worldwide coverage. The Differential GPS uses the NAVSTAR satellites and a beacon at a known site to reduce the error of the user position. The kinematic GPS system provides centimeter accuracy which can be corrected to the known landmarks when carrier phase ambiguity exists. Our work presents a highway environment simulation using our kinematic GPS model to provide three meter range accuracy for robotic applications. The three-dimensional problem for robotic vehicles is simplified by moving the vehicles on to a highway since the vehicle is constrained to move in one dimensional or a quasi-one-dimensional way. A detailed triple difference model for the GPS phase observable has been used to perform static positioning and kinematic positioning. Trimble data taken by Dr. Benjamin W. Remondi was provided for our use to simulate our model. It has been shown that the triple difference method achieves highly accurate results for the kinematic GPS positioning since the triple difference technique is insensitive to cycle slips. The algorithm for kinematic GPS positioning in a highway environment gives corrections to the receiver because the Over Head component of the error found in a highway environment could be removed and the Cross Track component could be resolved using stereoscopic sensors to detect the edge and middle of the road.