Font Size: a A A

Visual Recognition System Development For Position In The Process Of Grasping Parts On The Production Line

Posted on:2018-07-21Degree:MasterType:Thesis
Country:ChinaCandidate:W WangFull Text:PDF
GTID:2348330542460605Subject:Mechanical engineering
Abstract/Summary:PDF Full Text Request
At present,with the rapid progress of robot technology,the robot is widely used in the household,medical,industrial and other field.In particular,the extensive using of industrial robots promotes rapid development of China industry from the "big manufacturing country"to the "Intelligent manufacturing country".Now,most of the automatic production line of the standardization clamp and position parts with the accompanying fixture.After the parts arrive at the designated location,most of the industrial robots work through teaching or pre programming.In order to crawl and place the parts,robots must have precise point teaching or pre programming of its operation,which has significant limitations and less flexible.In order to change the disadvantages of the traditional teaching robot,this paper introduces the machine vision technology to identify the position and orientation of the disordered parts on the production line.Firstly,grasping the parts placed on the lathe accurately.then put the finished parts out on the temporary storage device after the lathe completes the "intelligent processing" of the production.Making the pump parts as an example,this paper mainly completes the work of "intelligent grasping".Firstly,design the overall scheme design of the visual recognition system of the parts on the production line.According to the requirements of the subject,the monocular vision system of robot eye-in-hand(eye in hand)is adopted,and the hardware and software were selected.Secondly,researches on the image processing technology of parts.Visual identification of the parts on the production line,pre processing is essential,such as image enhancement,image smoothing,edge contour analysis and detection.In order to achieve the accuracy requirements,several parts of the image are spliced into a high resolution panoramic image by image synthesis technology.Thirdly,research on the position and orientation of parts.According to the characteristics of the pump body,the value of the pump can be obtained from the inner circle and the circle of the water outlet.Form the two dimensional contour which is obtained by the edge fitting,and the method of determining the position of the pump body based on the center of the two circle is proposed.Fourthly,plan the trajectory of industrial robot.establishing the robot working space in RobotStuio,then using the RobotStuio built-in function to plan the robot joint trajectory and working the path planning and velocity analysis for each segment path,then getting the trajectory planning curve.Finally,robotstudio modeling and simulation.Built the static modeling and dynamic si mulation of the robot,conveyor belt,NC machine tool and workpiece grasping and placing.The dynamic simulation of the visual recognition of the position and posture in the process of the part of the production line is realized.
Keywords/Search Tags:robot, vision, image processing, pose, simulation
PDF Full Text Request
Related items