Epipole-based 3D visual servoing

This paper proposes a visual servoing algorithm for hand-eye robotic system based on epipolar geometry. The control law is based on the estimation of the epipoles position obtained by points correspondences extracted from the current and target images. The camera-robot motion is computed from the observation of the epipoles coordinates. Only the principal camera point is assumed to be known but not the other intrinsic parameters. Experimental results are reported to validate the visual servoing algorithm proposed.