A Human-Assisted Based Visual Servo Approach for Tele-Assembly

This paper proposed an image-based visual servo approach for a high precise tele-assembly task on TTRP (Nankai teleoperation/telegame robot platform). And it is proven that the visual servo approach developed by the paper guarantees stability of the position error asymptotically. Moreover, to enhance the robustness of the visual servo method for unstructured environment under the situation of coarse calibration, human assistance is involved to improve the precision of the image processing. To verify the validity of the method proposed by this paper, an experiment that two remote operators control two robot manipulators (6DOF) to accomplish a tele-assembly task with the precision less than 0.1mm is introduced. The experimental results are shown that the method proposed in this paper can improve the ratio of success effectively. Furthermore, through the experiments, the task can be accomplished only by two or three times´ regulation