Research On Welding Seam Detection And Robot Automatic Tracking Method Based On Monocular Vision Image

With the development of robot technology,welding industrial robot has a significant advantage in labor productivity and welding quality,so it has been widely quoted in all walks of life.Main idea of this article is based on weld image collection based on monocular vision,and robot automatic tracking,planning the path of the weld seam tracking system is designed,using machine vision,image processing and communication technology to realize the idea of this paper,the main work includes the following aspects:(1)Calibration of robot weld tracking systemAnalyzes the camera model of forming geometry,the camera distortion,as well as the analysis of robot and the camera Eye-in – hand system of coordinate system transformation relations,using software Halcon calibration of camera and hand-eye system.(2)The processing of weld images and planning of weld trajectoriesA series of image processing algorithms are used to obtain the trajectory planning of weld seam.First using the denoising image preprocessing based on Rough sets,get rid of the noise in the image using adaptive SUSAN algorithm for edge detection of weld,finally based on trajectory planning of weld seam in ridge line processing.(3)The design of the test bed for weld seam trackingSelect STEP 6 dof robot and it industry structures,seam tracking experiments based on monocular vision camera,and using the EtherCAT technology for data communication,do image processing and data transmission can be quickly.(4)Robot weld tracking experiment and analysisUse of test bench for trial operation,this paper introduces the steps and methods of the experiment,using the research methods of this article to weld seam tracking experiments,and analysis of the results of the experiment.