Robot self-location using visual reasoning relative to a single target object

摘要

This paper presents a computationally straightforward method for determining the location of a camera that is mounted on a robot. The procedure consists of observing a sphere upon which two great circles have been circumscribed and computing the location and orientation of the camera based upon projections of the primary features in the image plane. Distance from the standard mark is based on the size of the projected sphere, whereas location of the camera is based on the displacements of the great circles relative to the center of the projected sphere. The procedure for determining the location of the camera involves solving only linear equations and is computationally simple. Results demonstrating the utility of the method with both simulated data and actual images are presented.