Generation Method Based On The Soles Of The Structure Of Light Extraction And Robot Penjiao Track

With the improvement of living standards, people set higher requirements of their shoe leather such as the quality, comfort, personalized and so on, because high-quality leather shoes fit the individual foot force characteristics more. Then, the needs of individual shoes is gradually increasing. Recent years, the domestic shoe-making industry has developed rapidly, but the overall strength of big but not strong. At present, the shoe's sole coating process in a semi-automatic or manual operation level, has seriously affected the quality of leather shoes, personalized shoes even more out of the question. In this paper, the extraction of soles information and the generation of robot glue trajectory method were researched combines with the structured-light vision and industrial robotics, providing the basis theoretical of the shoe glue process automation, which will help to improve the shoe-making efficiency, enhance the quality of shoes, and promote the development of personalized shoes.Basic principles and perspective projection conversion model of structured-light vision testing were analyzed for linear structured light in the sole automatic glue application. On this basis, the target board has been made, and the X-corner's sub-pixel coordinates was obtained by using Harris corner point detection algorithm and the gray-square center of gravity method, and the camera's inside and outside parameters and the robot hand-eye relationship were calibrated in accordance with the camera's perspective projection model and the robot hand-eye model. The calibration accuracy is high enough, and the calibration error is less than 0.3 pixels. By using the calibrated parameters, the three-dimensional coordinates of the center points of sole light stripes were reconstructed, and the basic information about the point cloud data of shoe sole was obtained.In order to achieve automatic gluing of sole by robot, this paper presents a kind of new glue path generation method. Firstly, converted point cloud data into three-dimensional space scan line using B-spline fitting method, and then, extract all the endpoint of scan lines composed of point set Q, using B-spline curve interpolation algorithm generates sole contour. Finally, approximate calculate the bias curves of contours to obtain the sole glue track. The theoretical methods presented in the paper was achieved via VC++ 6.0, the system runs less than 6s to meet the requirements of real-time. Spray gun was controlled by St?ubli RX60 robot, discrete the trajectory line, calculate robot working points and location variables, and passed to the robot control program to realize the gun control movement. The running path basically consistent with the glue trajectory, which verifies the feasibility of this method.