Localization for Indoor Service Robot Using Natural Landmark on the Ceiling

Authors: Seung-Hun Kim, Changwoo Park

Abstract:

In this paper, we present a localization of a mobile robot with localization modules which have two ceiling-view cameras in indoor environments. We propose two kinds of localization method. The one is the localization in the local space; we use the line feature and the corner feature between the ceiling and wall. The other is the localization in the large space; we use the natural features such as bulbs, structures on the ceiling. These methods are installed on the embedded module able to mount on the robot. The embedded module has two cameras to be able to localize in both the local space and the large spaces. The experiment is practiced in our indoor test-bed and a government office. The proposed method is proved by the experimental results.

Keywords: Robot, Localization, Indoor, Ceiling vision, Local space, Large space, Complex space.

Digital Object Identifier (DOI): doi.org/10.5281/zenodo.1087830

References:

[1] C. Harris and M. Stephens (1988). "A combined corner and edge detector". Proceedings of the 4th Alvey Vision Conference. pp. 147–151.
[2] T. Lemaire, S. Lacroix, and J. Sola, "A practical 3D bearing-only SLAM algorithm," in Intelligent Robots and Systems, 2005. (IROS 2005). 2005 IEEE/RSJ International Conference on, 2005, pp.2449-2454.
[3] T. Lemaire, C. Berger, I.-K. Jung, and S. Lacroix,"Vision-Based SLAM: Stereo and Monocular Approaches," Int. J. Computer Vision, vol. 74, pp.343-364, 2007.
[4] Seung-Hun Kim, et al., " Self-Localization Embedded Module Based On Ceiling Segmentation Method For Mobile Robot" International Conference on Computer Engineering and Technology, 2012.
[5] Su-Yong An, et al., " Vision-Based Topological SLAM in Spacious Environment", ICROS, 2011, pp.149-151.