The Research Of Arm Grab System Based On Visual Guidance

Machine vision system calculates and processes the target digital image information collected by image acquisition device through image processing software,and returns the processing results.The field equipment completes the corresponding functions according to the results of image processing.The machine vision system has the advantages of high accuracy,high speed,non-contact and so on.It has the advantages of positioning guidance,recognition,measurement,appearance detection and other industries.The field has been widely used.This paper takes visual guidance as the research object,introduces visual guidance technology into traditional manipulator to achieve closed-loop system,so as to improve the intelligence of manipulator and promote the intelligence and flexibility of actual production line.This paper introduces the hardware composition of visual guided grabbing,and discusses the performance parameters of hardware devices in visual guided system and the design scheme according to actual requirements.Firstly,the influence of light source on image acquisition is analyzed,and how to obtain a high quality image is experimented.According to different camera hardware,an image acquisition system is developed based on.NET platform,which can collect images from mainstream cameras for secondary development.By studying the principle of camera aperture imaging,the corresponding relationship between image coordinate system and real space coordinate system is analyzed,and the distortion caused by lens is eliminated by checkerboard distortion correction method.In view of the difficulty in obtaining the coordinates of the manipulator in hand-eye calibration of visual-guided grabbing system,a method of hand-eye calibration of Yamaha manipulator based on least square method is proposed.The calibration error is 1.874.The algorithm is tested by visual alignment system.The maximum errors in X,Y and R directions are 0.12 mm,0.12 mm and0.26 degrees,respectively.The experimental verification shows that the method based on least square method is effective.The hand-eye calibration method of Yamaha manipulator has strong adaptability,simple implementation and meets the accuracy requirements.Then,the image collected by the camera is preprocessed,such as image graying and image filtering,which has reduced the interference of the change of external light source,thus reducing the complexity of the target location algorithm.In target localization,a combination of template matching rough localization based on feature extraction and fine localization based on line detection is proposed to locate the center position and angle information of the target more accurately.Finally,this paper takes the mobile phone screen as the grasping object of the manipulator.On the platform of.NET Famrawork 4.0,the software system of visual-guided manipulator grasping is developed.By simulating the actual industrial production environment,it is proved that the visual-guided manipulator grasping system achieves the desired effect.