Grasping Objects of Various Shapes by Calibration-Free Vision-Guided Robots.
Volker Graefe, Minh-Chinh Nguyen · 2000
An approach to vision-based control of calibrationfree robots is introduced. Similarly to living organisms, it uses an uncalibrated stereo vision system, but no world coordinates and no inverse pcrspective or kinematic transformations, and it enables an automatic adaptation to changing optical, mechanical and control parameters of the robot. We implemented the approach on a vision-guided calibration-free articulated arm manipulator with five degrees of freedom. In real-word grasping experiments we showed that robot control without calibration is, indeed, feasible and that it leads to robots with a rcmarkable degree of robustness against disturbances of the sensors and unexpected changcs of the robot's parameters. In this paper we describe how such a robot can be rcali~ed. I11 contrast to earlier implementations of our approach we now control all 5 degrees of freedom of the robot and, thus, enable the robot to manipulate a much larger variety of objects of different shapes (Figurc 1) and nearly arbitrary initial orientations. Related Work Object grasping without calibration has long occupied the attention of researchers in vision-based robot control. Recently a variety of methods has been developed for this challenging and difficult task. For example. [Hollinghurst, Cipolla 19941 and