Abstract:

Aiming at the seam automatic extraction and the stitching path planning of curve surface fabrics, a seam automatic recognition method based on binocular vision was proposed. Combined the spatial position conversion between three dimensional points cloud and the two dimensional image of a stitched prefabricated part, the map of the seam in two dimensional image of the prefabricated paart in the three dimensional points cloud was acquired. Furthermore, Freenet-Serret vectors in all positions of the stitching path were calculated by a path designing method based on Frenet-Serret vectors, the posture of a robot terminal was calculated, and the stitching path of the robot was simulated using MatLab. The experimental results show that the method can position the spatial positions of a seam accurately and quickly and solve the problems in the automatic stitching process on difficut stitching path search and low precision, and can also plan a stitching path of the robot quickly.

Key words: curve surface fabric, binocular vision, seam position, Frenet-Serret vector, stitching path planning