Analyzing accuracy of GPS data for vehicular parameters

Currently, most vehicular parameters, such as speed acceleration and heading, are obtained from GPS data. In this paper our goal is to investigate an alternate source of vehicular information for collision avoidance systems and driver assistance applications, which is more accurate, reliable and has significantly lower time lag. We have implemented an architecture, which enables us to read, analyze, decode and store the real-time vehicular data from the vehicle´s electric sensors. We have also presented and analyzed experimental data showing the errors in vehicular speed obtained from GPS readings, under different driving conditions.