Design And Research On Transport System Based On 6-axis Industrial Robot And Vision Positioning

With the transport robot' widely using in industrial production,the traditional teaching methods can not work when the pose of the object to be carried is not bound.To solve the problem,vision technology is introduced to recognize and locate the target object,guide the robot to complete the task of conveying.Taking air-conditioning compressor onto production line as the objective task,robot kinematics,camera calibration and hand-eye calibration,image processing and target recognition are all involved in the research.Besides,in the last part of the paper,the transport robot based on the vision positioning is designed and applied,according to the task requirements.In the research of robot kinematics part,based on the structure and basic parameters of RS50 N type robot,the D-H coordinate system is established,corresponding D-H parameters are listed,and the forward&inverse solution of kinematics is obtained,which provide technical support for the hand-eye calibration and vision positioning of the transporting system.In the research of camera calibration part,the basic mathematical model of camera is discussed,then camera calibration experiment is conducted by Zhang Zheng-you calibration method based on Matlab camera calibration toolbox,the intrinsic parameter and the corresponding external parameter of the camera are derived;In the research of hand-eye calibration part,considering the Eye-in-Hand vision system,the basic relation AX = XB of eye-hand calibration is derived,the solution is conducted by Kronecker product method.Further,total least squares method is introduced,in order to reduce the impact of noise.Finally,eye-hand calibration of Eye-in-Hand system is carried out by the experiment.In the research of image processing and target recognition part,taking Matlab image processing toolbox as the platform and an image captured as the processing object,image preprocessing and the edge detection technique is researched;Aim at circular object recognition in image,giving a new method combined Circle Hough Transform and least square method.The experimental results show that the method is effective with high recognition efficiency and positioning accuracy.Finally,according to the above research results and the task requirements,auto transporting system for handling air-conditioning compressor is built based on machinevision and RS50 N robot.Thereinto,the gripper mechanism installed in robot end is designed and optimized.The experimental result shows that the system can work well.