The accuracy of 6D SLAM using the AIS 3D laser scanner
Automatic sensing of the environment is a fundamental scientific issue in robotics, since it is essential for autonomous mobile robot systems. In previous works, we presented a 6D SLAM algorithm which is based on the spatial data from the AIS 3D laser scanner and a variant of the iterative closest points algorithm (ICP). In this paper we focused on the reachable accuracy of the whole approach and therefore performed several ground truth experiments. We will show that the 6D SLAM algorithm can compensate for erroneous data of the 3D laser scanner, at least in a limited range. Furthermore we will discuss different aspects which influences the accuracy of our approach.