Aperture detection and alignment control based on structured light vision

Apertures play a critical role as the marks for workpiece positioning on assembly lines. The detection and alignment of an aperture on workpiece surface is a typical manipulation task for assembling robots. In this paper, a vision sensor based on laser structured light is presented to detect an aperture and align it to a robotic end-effector with an eye-in-hand vision system. The images with laser stripes are segmented via adaptive threshold and the stripes skeleton are extracted, then the cross stripes are detected by Hough transformation. The feature points of the aperture are extracted according to intensity distribution of pixels on the cross. A set of visual features and a partitioned visual control law are presented for aperture alignment based on structured light. Experiments have been conducted to test the effectiveness of the algorithms developed.