Localization for Autonomous Vehicle on Urban Roads
Open Access
Abstract: Autonomous driving is an emerging technology in which a car performs recognition,
decision making, and control in place of a human driver. Localization is one of the core issues for
autonomous driving. In particular, driving on urban roads requires highly dependable localization
techniques, because the traffic environment is more complicated than that of an expressway. In this paper, we propose a localization algorithm for an autonomous vehicle, which estimates the vehicle’s position by means of template matching. The results show that this algorithm is better suited to make estimations in various environments than the Global Navigation Satellite System and Inertial Measurement Unit (GNSS/IMU) system. Specifically, the method used in this study enables autonomous operation the vehicle up to 13km on public roads, while the GNSS/IMU system cannot achieve this.
Keywords: autonomous vehicle; localization; LIDAR; infrared reflectivity; template matching
