Simultaneous Localization and Mapping with the Kinect sensor

Konferenz: ROBOTIK 2012 - 7th German Conference on Robotics
21.05.2012-22.05.2012 in Munich, Germany

Tagungsband: ROBOTIK 2012

Seiten: 6Sprache: EnglischTyp: PDF

Persönliche VDE-Mitglieder erhalten auf diesen Artikel 10% Rabatt

Emter, Thomas (Fraunhofer Institute of Optronics, System Technologies and Image Exploitation IOSB, Karlsruhe, Germany)
Stein, Andreas (Robert Bosch GmbH, Reutlingen, Germany)

A very important prerequisite for mobile robots to navigate autonomously is their ability to build a map while exploring unknown areas. In this paper we present an algorithm for simultaneous localization and mapping for indoor environments based on the Kinect sensor. The algorithm is capable of building a 3D voxel map including color information and performing localization with scan matching. A great reduction of the run-time has been achieved with the RANSAC-method by choosing a subset of points for scan matching compared to the full set without compromising the precision of the mapping. Conference Topic: 4. Robotics in Service / Mobile Robotics Keywords: autonomous mobile robot, SLAM, Kinect, scan matching, RANSAC