G-ICP SLAM: An odometry-free 3D mapping system with robust 6DoF pose estimation

Abstract

The paper proposes an odometry-free 3D mapping system that combines a LIDAR and a inertial sensor. The proposed system achieved robust 6DoF pose estimation for arbitrary motion and is implemented as a hand-held unit to make use of simplified mobile mapping applications. The pose estimation algorithm is based on "Velodyne SLAM" which is a state of the art ICP based SLAM (Simultaneous Localization and Mapping) method using only point cloud data. We added 3DoF inertial information to process the point cloud correction and the position prediction. Compared to the previous method, the proposed method is robust to rotary motion and works for fast and large change of sensor position and orientation. The results demonstrate effective operation in various environments and we confirmed the improvement of the self-position estimation and mapping performance.