nextuppreviouscontents
Next:Multi-Robot Collaboration Up:Active Vision Previous:Virtual Environment Construction

Accurate Position Estimation from Learned Visual Landmarks

R. Sim, G. Dudek

This method for learning, encoding, detecting, and using visual landmarks for mobile robot pose estimation. The work deals primarily with landmarks from monocular video cameras and can be used to learn a small set of robust landmarks to localize a robot (or other device) in an almost-arbitrary environment.
 


Annual Report

Fri Nov 26 23:00:32 GMT 1999