A dense structure model for image based stereo SLAM

Author

Tykkälä, Tommi ; Comport, Andrew I.

Author_Institution

CNRS, UNSA, Sophia-Antipolis, France

fYear

2011

fDate

9-13 May 2011

Firstpage

1758

Lastpage

1763

Abstract

In this paper a dense structure model is developed for stereo image based Simultaneous Localization And Mapping (SLAM). It is proposed to model dense environment structure incrementally by robustly integrating disparity maps from current and previous time instants. In this way disparities can be refined over time to favor consistent 3D structure over noise. The analytical search bounds for disparities are transferred into the current map to allow efficient re-localization. The cost function is image-based and it is minimized by combining Iteratively Reweighted Least-Squares (IRLS) with exhaustive search for finding motion and disparity parameters respectively.

Keywords

SLAM (robots); iterative methods; least squares approximations; mobile robots; robot vision; stereo image processing; 3D structure; dense structure model; disparity parameter; image based stereo SLAM; iteratively reweighted least square; model dense environment structure; motion finding; robustly integrating disparity map; simultaneous localization and mapping; Accuracy; Cameras; Optimization; Rocks; Simultaneous localization and mapping; Three dimensional displays; Trajectory;

fLanguage

English

Publisher

ieee

Conference_Titel

Robotics and Automation (ICRA), 2011 IEEE International Conference on

Conference_Location

Shanghai

ISSN

1050-4729

Print_ISBN

978-1-61284-386-5

Type

conf

DOI

10.1109/ICRA.2011.5979805

Filename

5979805