Parallel Robot Welding Seam Tracking System Based On Laser Vision

Welding is one of the main material processing technologies in the industry,whether it is some common small-scale furniture products in life,or large-scale machinery and equipment used in the production and use of enterprises,or in the world’s cutting-edge aerospace and ship fields are closely related to welding..Compared with serial robots,parallel robots have the advantages of light structure,fast dynamic response,and high precision,and can meet the needs of high precision in the welding field.Aiming at the problem that the parallel robot is easy to move but difficult to realize flexible welding,the visual sensor is used to provide deviation information,and the position and posture of the robot are adjusted.Therefore,the parallel welding robot is used for welding experiments,combined with laser vision sensors,to track,correct and detect the deviations generated by the parallel robot technology in the actual welding process in real time.The specific research contents are as follows:(1)The structure types and kinematics of parallel robots are studied.By analyzing the interconnection and mathematical relationship between the robot components,the kinematics solution equation with the visual acquisition deviation signal is established,and the action required at the end of the robot is judged,and the inverse solution equation model of the experimental fuzzy control system is obtained.(2)Through the detection principle of the laser vision sensor,the information extraction of the welding groove feature is realized by the maximum distance method.The information collected by the sensor is processed by the median filter method,a deviation correction algorithm is proposed to obtain the deviation information,and the hand-eye calibration is performed to obtain the experimental hand-eye transformation matrix,and finally the actual deviation data of the welding torch in the XYZ directions are obtained.(3)Design a two-dimensional fuzzy controller with welding torch position deviation E and deviation change rate EC as input variables and control quantity u as output,fuzzy processing E and EC,define fuzzy rules,and use the center of gravity method to clarify the fuzzy quantities.The rotation angle output by the controller is compared with the rotation angle obtained from the inverse solution,and input to the Delta robot model to calculate the position coordinate point.Compare this coordinate point with the deviation coordinate to get the deviation data.Decoupling the multi-input multi-output fuzzy control system,using the deviation data ΔX,ΔY,ΔZ information input,through the multi-input and multi-output fuzzy control system to obtain the real-time tracking of the four joints of the parallel robot.(4)A parallel robot laser vision welding experimental platform was built and welding experiments were carried out.The results show that the proposed method can be positioned accurately,and the overall process performance of each weld test can meet the technical requirements,which not only meets the expectations of the weld test weld test,but also proves the practicability of the designed welding tracking control system.