Homography-based visual control of nonholonomic vehicles: A new approach

Abstract

This paper presents a new visual control approach based on homography. The control law is designed for nonholonomic vehicles with a fixed monocular system on board. The method uses a homography measured from the reference image and the current image in the feedback control law. The idea of the visual control used here is the usual approach where the desired position of the robot is given by a target image taken at that position. When starting from an unknown position the control law uses the feedback from the camera to move the robot towards this target image plane which is the only previous information needed to position the robot at the desired position. The contribution is the fuzzy controller used as a visual feedback control that is an intelligent controller without needing neither the decomposition of the homography nor the depth estimation to the target.