MEASUREMENT AND STORAGE OF A NETWORK OF JACOBIANS AS A METHOD FOR THEVISUAL POSITIONING OF A ROBOT ARM

The goal of this paper is to describe a method to position a robot arm at any visible point of a given workspace without an explicit on line use of the analytical form of the transformations between real space and camera coordinates (camera calibration) or between cartesian and j oint coordinates (direct or inverse kinematics of the robot arm). The formulation uses a discrete network of points distributed all over the workspace in which a procedure is given to measure certain Jacobian m atrices which represent a good local linear approximation to the unkno wn compound transformation between camera and joint coordinates. This approach is inspired by the biological observation of the vestibule-oc ular reflex in humans (VOR). We show that little space is needed to st ore the transformation at a given scale, as feedback on The visual coo rdinates is used to improve precision up to the limit of the visual sy stem. This characteristic also allows the plant to cope with disturban ces in camera positioning or robot parameters. Furthermore, if the dim ension of the visual space is equal or bigger than the motor space dim ension, the transformation can be inverted, resulting in a realistic m odel of the plant able to be used to train other methods for the deter mination of visuo-motor mapping. As a test of the method an experiment to position a real robot arm is presented, together with another expe riment showing the robot executing a simple task (building a tower of blocks).