The robot, the scanner and the turntable form a scanning platform that we use to obtain 3-D models of real objects placed on the turntable. Since objects come in different shapes, it is necessary to scan them from many different viewpoints. This is achieved by turning the turntable. In order to stitch the scans we obtain this way into a coherent single 3-D model we need to have an accurate estimate of the relative positions and orientations of the laser scanner and the turntable.
This project investigates the issue of how to automatically determine the needed transformation.