Research On 6 Dof Robot Welding Control Based On Visual

In recently years, with the development of society and the advancement of technology, industrial production has a rapid development. Industry represents an important economic pillar of the nation, which is also an important symbol to measure a country’s comprehensive national strength. With the improvement of industrial production requirements, the product of high-efficiency, high-quality and high-precision has become an inevitable trend. The emergence of welding robot has changed the structure of the industry, especially in the manufacturing industry.Machine vision has become an important technology to ensure the efficiency and quality in the process of industrial production. The technology of welding robot control based on vision has the features of automated, intelligent and informative.This article focuses on the research of the weld image processing or recognition and the robotic welding control.Firstly, the current condition of the vision-based welding robot is introduced briefly, and the two categories of robot vision system is explained detailed. The paper has made a detailed description of camera models, including the principle of pinhole imaging. It also establishes a model of binocular vision, and thus deduces the image Jacobian. In addition, it also makes a brief description of the robot kinematics and drives the robot Jacobian.Secondly, the processing of weld image based on visual is described detailed. In the median filtering process, the article uses an adaptive median filter to improve filtering performance in order to improve the traditional methods. In order to reduce the image processing time and improve the speed of image processing, the article only gets the interested part of the image. In the image segmentation process, an adaptive method to determine the threshold value of binarization is used in this article. This article gets the weld feature points by the way of slope changing. And it detects the band of structure light using Hough transform.Finally, an uncalibration visual servo controller is used to deal with the positioning control of the welding robot gun traction to the desired position be welded. The article uses adaptive estimation of parameters and the image errors is asymptotic convergence to zero by the lyapunov theory. Simulation is demonstrate the effectiveness of the proposed scheme and provides the theoretical preparation for the later experiment.