Title :
Real-time image-based topological localization in large outdoor environments
Author :
Bradley, David M. ; Patel, Rashmi ; Vandapel, Nicolas ; Thayer, Scott M.
Author_Institution :
Robotics Inst., Carnegie Mellon Univ., Pittsburgh, NY, USA
Abstract :
This paper presents a real-time implementation of a topological localization method based on matching image features. This work is supported by a unique sensor pod design that provides stand-alone sensing and computing for localizing a vehicle on a previously traveled road. We report extensive field test results from outdoor environments, with the sensor pod mounted on both a small and a large all-terrain vehicle. Off-line analysis of the approach is also presented to evaluate the robustness of the various image features tested against different weather and lighting conditions.
Keywords :
feature extraction; geographic information systems; image matching; navigation; sensors; all-terrain vehicle; image feature matching; real-time image-based topological localization; sensor pod; Cameras; Data mining; Image analysis; Indoor environments; Instruments; Large-scale systems; Road vehicles; Robot sensing systems; Robustness; Testing;
Conference_Titel :
Intelligent Robots and Systems, 2005. (IROS 2005). 2005 IEEE/RSJ International Conference on
Print_ISBN :
0-7803-8912-3
DOI :
10.1109/IROS.2005.1545442