Research On Robot Glue Loading System Based On Vision

The traditional printing binding process adopts manual feeding method,and there are many drawbacks in the working process,such as low production efficiency,harmful substances in the rubber material and noise generated by the binding machine when working will damage the health of workers,and the increasing labor cost leads to a continuous decline in profits in the printing industry.In order to solve these problems,this thesis studies the vision-based robotic arm glue feeding system,and mainly completes the following aspects:(1)The forward and reverse kinematics of the robotic arm were analyzed,the D-H modeling method was used to model the kinematics of the AUBO-E5 manipulator,and the simulation experiment was carried out using the robot toolbox in the MATLAB software platform to verify the correctness of the positive and reverse solution of the AUBO-E5manipulator;(2)The industrial camera was selected and the calibration method of the camera was discussed.Firstly,it is determined that Zhang Zhengyou calibration method is selected to complete the internal parameter calibration of the camera.Then select the eye-to-hand installation mode,and use the Aruco two-dimensional code notation method to perform handeye calibration to complete the coordinate transformation between the camera and the coordinate system of the robotic arm;(3)The method of combining teaching reproduction and visual servo was adopted to complete the feeding path planning of the robotic arm.(4)The images captured by the camera are processed to achieve precise positioning and loading.Firstly,the average method is used to transform the image in grayscale,the median filtering method is selected to denoise the image,and the Otsu(Otsu method)is used to calculate the image threshold to complete the threshold segmentation of the image.Secondly,the contour of the stapling groove and the contour of the paper to be loaded are extracted features,and then a geometric algorithm for paper and stapling groove is proposed according to their position relationship,so that the paper can be accurately placed into the stapling groove.In this thesis,an experimental platform of robotic glue binding feeding system imitating the real scene was finally built,and papers feeding experiments were carried out on it.The experimental results show that the system can grasp some papers according to the planned path,and the visual servo technology is applied to complete the feeding work with high precision in the eye-in-hand mode.