Abstract
SLAM mechanisms are a key component towards advanced service robotics applications. Currently, a major hurdle are the still high costs of suitable range measuring devices. A solution are bearing-only SLAM approaches since these can be used with cheap sensors like omnicams. The general approach of using an Extended Kaiman Filter (EKF) for bearing-only SLAM based on artificial landmarks has been described and evaluated in [2][1]. Instead of artificial landmarks, we now use SIFT features [3] as natural landmarks.
This paper describes SIFT feature preselection and landmark identification mechanisms that are pivotal towards the robust application of SIFT features within a bearing-only SLAM approach based on the EKF. We exploit viewing areas to massively reduce ambiguities and mismatches in SIFT feature reobservations and thus significantly reduce false identifier assignments.
The approach has been successfully evaluated on a Pioneer-3DX platform in an unmodified indoor environment. The results show that bearingonly SLAM produces reliable results even with cheap vision sensors and natural landmarks.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
Similar content being viewed by others
References
C. Schlegel and S. Hochdorfer. Bearing-Only SLAM with an Omnicam — An Experimental Evaluation for Service Robotics Applications. 19. Fachgespräch Autonome Mobile Systeme (AMS), Stuttgart, 99–106, Springer, 2005.
T. Bailey. Constrained Initialisation for Bearing-Only SLAM. In: Proc. IEEE Int. Conf. on Robotics and Automation (ICRA), pages 1966–1971, Taipei, Taiwan, 2003.
D. Lowe. Distinctive image features from scale-invariant keypoints. International Journal of Computer Vision, 60: 91–110, 2004.
J. Hesch and N. Trawny. Simultaneous Localization and Mapping using an Omni-Directional Camera. Available: www-users.cs.umn.edu/∼joel/_files/BoS.pdf.
H. Tamimi, H. Andreasson, A. Treptow, T. Duckett, A. Zell. Localization of mobile robots with omnidirectional vision using Particle Filter and iterative SIFT. Robotics and Autonomous Systems 54(9): 758–765 (2006)
A. Gil et al. Simultaneous Localization and Mapping in Indoor Environments using SIFT Features. In: Proc. of the IASTED Conf. on Visualization, Imaging, and Image Processing (VHP). Palma de Mallorca, Spain, 2006.
A. J. Davison et al. MonoSLAM: Real-Time Single Camera SLAM. In: IEEE Transactions on Pattern Analysis and Machine Intelligence. Vol. 29, No. 6, pages 1052–1067, June 2007.
Further experimental results: http://www.hs-ulm.de/schlegel, then follow Bearing Only SLAM
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2007 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Hochdorfer, S., Schlegel, C. (2007). Bearing-Only SLAM with an Omnicam Robust Selection of SIFT Features for Service Robots. In: Berns, K., Luksch, T. (eds) Autonome Mobile Systeme 2007. Informatik aktuell. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-74764-2_2
Download citation
DOI: https://doi.org/10.1007/978-3-540-74764-2_2
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-74763-5
Online ISBN: 978-3-540-74764-2
eBook Packages: Computer Science and Engineering (German Language)