Study On Weld Seam Recognition And Trajectory Planning Method Of Industrial Robot

As the support technology and important automation equipment of modern manufacturing industry,industrial robots have been introduced in many areas of intelligent manufacturing.Especially in the field of welding,the traditional manual operation has gradually been replaced by industrial robots.The existing industrial robot welding system has some disadvantages,such as tedious preliminary preparation,insufficient external sensing ability,lack of autonomous operation ability and so on,which have certain limitations.With the rapid development of machine vision,it can improve the sensing ability of the robot to a certain extent by introducing it into the welding scene of industrial robot,which is of great significance to improve the intelligence level of welding operations and even the manufacturing industry.This thesis takes the operation of industrial robots in the welding scene as the background,and aims to improve the autonomous operation ability and intelligence level of the industrial robot's welding operations.and on the theoretical basis of robot kinematics,machine vision and trajectory planning,the methods of weld seam information processing and robot trajectory planning in Cartesian space are studied.The design of the welding system is completed to verify the effectiveness of the theory in this thesis.The work of this thesis mainly includes the following:(1)Scheme designed of welding system for industrial robot.The controller,robot,welding scheme and detection device that make up the system are demonstrated and designed,and the basic scheme of the system is determined.Aiming at the constructed robot binocular Eye-to-Hand visual inspection system,this thesis introduces its imaging principle,the principle and method of monocular and binocular calibration,hand eye calibration,which provides the basis for the weld seam information processing and welding torch positioning.Aiming at the industrial robot,the D-H method is used to complete the kinematics modeling of the robot,and the forward and inverse kinematics analysis is performed to provide the basis for the trajectory planning and control of the robot welding operation.(2)Study on image processing method of welding seam information.The preprocessing algorithm for image processing of weld seam information is introduced.The edge detection of weld seam is researched using Canny edge detection algorithm.The center line extraction of weld seam is researched using Zhang-suen thinning algorithm.According to the principle of binocular stereo vision and epipolar constraint,the stereo matching of the center line of the weld is studied,and the three-dimensional coordinate measurement of the weld is realized.(3)Study on trajectory planning algorithm of industrial robot.Aiming at the welding operation of curved weld seam,the trajectory planning algorithm of robot in Cartesian space is studied.The algorithm of segmented path planning based on PSO is studied,and completed the simulation verification in Matlab.The algorithm of velocity planning based on sine acceleration curve and cross-section look-ahead is studied,and completed the simulation verification in Matlab(4)Software and hardware designed and verified of robot welding system.Based on the system scheme,the system of industrial robot is further achieved,which mainly includes the model selection of the hardware part and the program design of the software part.In order to verify the effectiveness of the weld seam information processing and robot trajectory planning algorithms in practical application,the camera calibration experiments,welding information processing experiments,welding torch positioning experiments and welding experiments were Designed.The maximum errors between the coordinates calculated by the visual recognition and the measured coordinates in the X,Y and Z directions are all less than 0.32mm.The welding experiments effect is uniform and continuous,and the welding effect meets the system requirements.The experimental results show that the method in this thesis can effectively improve the autonomous operation ability and intelligence level of industrial robots in welding scenarios.