Bachelorarbeit BCLR-2016-48

Schlüter, Henning: 3D-Mapping using Planar Surface SLAM on Mobile Robots.
Universität Stuttgart, Fakultät Informatik, Elektrotechnik und Informationstechnik, Bachelorarbeit (2016).
73 Seiten, englisch.
CR-Klassif.I.2.1 (Applications and Expert Systems)
I.2.9 (Robotics)
I.4.6 (Image Processing and Computer Vision Segmentation)

A feature-based real-time mapping approach using bounded planar surfaces extracted from depth images (like those produced by 3D depth cameras) will be presented in this paper. The mapping system was devised with the intention of utilising it for active autonomous exploration employing multiple mobile robots. The system itself is based on factor-graph optimisation using iSAM2. A representation for robot poses and plane landmarks as variables in such a graph is provided. Then constraints resulting from measurements, i.e. odometry and extracted planes, between the variables are derived as factors in the graph. This results in two different factors for planes, which are then analysed in experiments. The modular system developed to handle the data and manage the factor-graph is capable of being expanded to handle further types of input measurements and multiple robots.

Volltext und
andere Links
PDF (966850 Bytes)
Zugriff auf studentische Arbeiten aufgrund vorherrschender Datenschutzbestimmungen nur innerhalb der Fakultät möglich
Abteilung(en)Universität Stuttgart, Institut für Parallele und Verteilte Systeme, Maschinelles Lernen und Robotik
BetreuerToussaint, Prof. Marc
Eingabedatum26. September 2018
   Publ. Informatik