High-speed spatial change detection using point clouds and NDT

Abstract

This research aims to develop a high-speed spatial change detection technique using point clouds and NDT (Normal Distribution Transformation). Point cloud obtained from a range sensor such as a RGB-D camera is transformed to voxel representation using NDT, and aligned and compared with map data measured by a high-precision laser scanner. Three techniques are introduced to make the proposed system robust for noise, that is, classification of point distribution, overlapping of voxels, and voting using consecutive sensing.