Dense 3D Map Building for Autonomous Mobile Robots

초록

Numerous approaches for vision-based navigation of mobile robots using depth maps are classified into two classes: the first class is to extract image features, such as edge segments, and produce sparse depth maps, and the second class is based on correlation algorithms to produce more dense depth maps. One of the problems of the first class is that it may not offer enough information of the scene for autonomous navigation. In this paper, we use the second class algorithm to build more precisely a 3D map, not a depth map, and extract 3D visual cues for autonomous navigation.