A simple approach for 2D visual servoing

Abstract

Visual servoing is a useful approach for robot control. It is specially attractive when the control objective can be stated directly in image coordinates like in point-to-point regulation. In this work, we proposed a control algorithm which allows to specify the desired final position through the computer screen and which requires for implementation only joint and image coordinates, while a velocity observer for the last ones is designed. Since a map between both coordinates is not employed, the knowledge of camera parameters is not necessary. This is achieved by adapting a control algorithm designed originally to work in Cartesian coordinates to a version which uses image coordinates instead. It is assumed that the robot is planar and the camera fixed, so that the image plane is parallel to the manipulator workspace. Experimental results show the good performance of the complete system.