Change search
ReferencesLink to record
Permanent link

Direct link
Indoor Visual Localization of the NAO Platform
KTH, School of Computer Science and Communication (CSC).
2013 (English)Independent thesis Advanced level (degree of Master (Two Years)), 20 credits / 30 HE creditsStudent thesisAlternative title
Lokalisering i inomhusmiljö med NAO-plattformen (Swedish)
Abstract [en]

Global localization in an indoor environment using the NAO humanoid robot has been studied and tested experimentally. The solution relies on a pre-existing 3D map consisting of so called SURF features. Using the NAO built-in camera, SURF features are extracted and matched against the 3D map. The found 2D-3D correspondences are used to solve the PnP (Perspective-n-Point) problem and thus provide a measurement of the camera pose. These measurements are fused with odometry measurements in an Extended Kalman Filter, to provide a final estimate of the robot pose. The algorithm has been implemented and tested with a real NAO robot, with reference measurements provided by an external positioning system.

Abstract [sv]

Global lokalisering av NAO-robotar har studerats och testats experimentellt. Lösningen bygger på en tidigare skapad 3D karta, bestående av s.k. SURF beskrivningar. Ur bilder tagna med NAO-robotens inbyggda kamera extraheras SURF punkter och dessa matchas sedan mot 3D kartan. Matchningarna mellan bild och karta (2D till 3D) används för att ta fram en mätning av robotens position. Ett Extended Kalman Filter används för att kombinera dessa mätningar med odometri från roboten, vilket resulterar i en slutgiltig estimering av robotens position. Metoden har implementerats och testats med en riktig NAO-robot, tillsammans med ett externt positioneringssystem för referensdata.

Place, publisher, year, edition, pages
National Category
Computer Science
URN: urn:nbn:se:kth:diva-142421OAI: diva2:700304
Educational program
Master of Science in Engineering - Computer Science and Technology
Available from: 2014-03-11 Created: 2014-03-04 Last updated: 2014-03-11Bibliographically approved

Open Access in DiVA

fulltext(5247 kB)304 downloads
File information
File name FULLTEXT01.pdfFile size 5247 kBChecksum SHA-512
Type fulltextMimetype application/pdf

By organisation
School of Computer Science and Communication (CSC)
Computer Science

Search outside of DiVA

GoogleGoogle Scholar
Total: 304 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

Total: 204 hits
ReferencesLink to record
Permanent link

Direct link