An auto-epipolar strategy for mobile robot visual servoing

Abstract

A novel visual servoing method is presented. The algorithm works for both apparent contours and point features and does not require any information about the internal camera parameters. The proposed visual servoing is based on the epipolar geometry and exploits the autoepipolar property, a special configuration for the epipoles which occurs when the desired and current views undergo a pure translation. This occurrence is simply detectable from special line conditions on the overlapped current and desired images. Experiments are presented to validate the proposed visual servoing.