| Industrial robotics technology is widely used in all aspects of industrial production and plays an important role in equipment manufacturing.At present,the industrial robots working on the automatic production line mainly control the robots to perform point-to-point tasks according to the predetermined trajectory by instructing or off-line programming.This control mode requires the accurate positioning of workpieces by tooling fixtures and the precise coordination between the industrial robots and the workpiece transfer system,which requires a high degree of automation of the production line.This greatly limits the application range of industrial robots.In order to solve the limitations of traditional robotic control methods,this research applies machine vision technology to industrial robotic positioning,uses machine vision technology to identify and position workpieces online,assists to guide industrial robots to complete tasks such as grasping,handling and assembling,so as to improve the flexibility in the application process of industrial robots,and has good practical application value.Firstly,the composition of visual positioning and grabbing system for industrial robots is introduced.Starting from the general frame structure of machine vision system,three different types of machine vision system are studied,and the overall design of the system is carried out to meet the requirements of experiment and use.The hardware composition and software development environment of the system are selected and the workflow of the system is introduced.Secondly,the parametric model of visual positioning and grabbing system of industrial robots is established.The principle of coordinate transformation and the position and attitude transformation among links of industrial robots are studied to establish the coordinate transformation relationship during the process of robot grasping.The camera imaging model and robot eye model are studied.Zhang Zhengyou calibration method is used to calibrate the camera internal parameters,to construct the mapping of two-dimensional pixel coordinates of workpiece image to real-world coordinates,to calibrate the eye system by Tasi two-step method,and to establish the transformation relationship from camera coordinate system to robot basic coordinate system.Then,target recognition and location algorithm of workpiece are studied.This paper mainly carries out theoretical analysis and practical verification on each link of image preprocessing,so as to select a suitable preprocessing scheme.Afterwards,target recognition algorithm is determined by analyzing the feature parameters of workpiece(contour area,ratio of length to width of minimum external rectangle,ratio of contour area to minimum external rectangle area,Hu moment).Finally,the position of workpiece is positioned by means of camera coordinate transformation relationship and camera calibration internal reference,and the stability and accuracy of the algorithm are verified by experiments.Finally,the above algorithm and the whole system are tested and analyzed.The experiment and error analysis of industrial robot positioning based on vision positioning are carried out on the experimental platform.The maximum Euclidean distance between real coordinates and vision positioning coordinates is 1.58 mm.Through experiments,the system meets the requirements of this project.Based on the support of basic scientific research projects of National Defense Science and Technology Bureau,National Natural Science Foundation of China and Science and Technology Program of Inner Mongolia Autonomous Region,this paper researches the visual positioning method of industrial robots grabbing by machine vision,focusing on the workpiece recognition and positioning requirements in the process of automatic handling of industrial robots.An industrial robot visual positioning system experimental platform is built with KUKA industrial robot and RVC X industrial camera as hardware platform,and the accuracy of the visual positioning method is verified by the experiment. |
PDF Full Text Request