3D model based map building

Comunicación presentada en el IX Workshop de Agentes Físicos (WAF'2008), Vigo, 11-12 septiembre 2008.

Several works deal with 3D data in SLAM problem. Data come from a 3D laser sweeping unit or a stereo camera, both providing a huge amount of data. In this paper, we detail an efficient method to extract a 3D model from raw data. Each model includes planar patches that belong to planar surfaces on the real scene. Then, we use these models in an ICP-like method in order to compute the robot movement. Once we know the movement performed by the robot, we can build a 3D environment map. Using ICP with planes is not a trivial task. It needs some adaptation from the original ICP. Some promising results in 6DoF are shown for both indoor and outdoor environment and for different kinds of 3D sensors.

This work has been supported by project DPI2005-01280 from Ministerio de Educación y Ciencia (Spain).