A few groups use 3D laser scanners [1,5,11,14,15]. The RESOLV project aimed to model interiors for virtual reality and tele presence [11]. They used a RIEGL laser range finder on robots and the ICP algorithm for scan matching [3]. The AVENUE project develops a robot for modeling urban environments [1], using an expensive CYRAX laser scanner and a feature-based scan matching approach for registration of the 3D scans in a common coordinate system. Nevertheless, in their recent work they do not use data of the laser scanner in the robot control architecture for localization [5]. Triebel et al uses a SICK scanner on a 4 DOF robotic arm mounted on a B21r platform to explore the environment [14].
Instead of using 3D scanners, which yield consistent 3D scans in the first place, some groups have attempted to build 3D volumetric representations of environments with 2D laser range finders [7,8,13,15]. Thrun et al. [7,13] use two 2D laser range finder for acquiring 3D data. One laser scanner is mounted horizontally, the other vertically. The latter one grabs a vertical scan line which is transformed into 3D points based on the current robot pose. The horizontal scanner is used to compute the robot pose. The precision of 3D data points depends on that pose and on the precision of the scanner. Howard et al. uses the restriction of flat ground and structured environments [8]. Wulf et al. let the scanner rotate around the vertical axis. They acquire 3D data while moving, thus the quality of the resulting map crucial depends on the pose estimate that is given by inertial sensors, i.e., gyros [15]. In this paper we let rotate the scanner continuously around its vertical axis, but accomplish the 3D mapping in a stop-scan-go fashion, therefore acquiring consistent 3D scans as well.
Other approaches use information of CCD-cameras that provide a view of the robot's environment. Some groups try to solve 3D modeling by using a planar SLAM methods and cameras, e.g., in [4].