Detection of road boundaries and obstacles using LIDAR

This paper proposes a novel approach to detecting road boundaries and obstacles for robust urban navigation of unmanned ground vehicles (UGVs). In this method, a 2-D Light Detection and Ranging (LIDAR) sensor mounted at a given pitch angle is used to extract the road surface. The information obtained about the road is evaluated by filters to detect whether the area scanned is a road or an obstacle. The proposed method is implemented through simulation and a real-life test vehicle called AKAY01.