3D point cloud map construction based on line segments with two mutually perpendicular laser sensors

Abstract

This paper presents a 3D point cloud map construction method based on extracted line segments with two mutually-perpendicular laser sensors in unknown indoor environment. To correct the position of mobile robot, the line segments are extracted from the raw sensor data from a horizontally installed laser sensor. In each step, these extracted segments are chosen as landmarks and matched with the stored features in previous steps. The state of mobile robot is updated by the successful matched line segments. Raw sensor data obtained from the vertical laser sensor are used to construct the 3D point cloud map with the small displacement of mobile robot. The experiment result of constructed 3D map of an unknown environment shows the feasibility of proposed method. © 2013 IEEE.