3D Visual Guidance And Alignment System For Flexible Assembly Of Mobile Phone Frame Screw Holes

The rapid development of the 3C electronic manufacturing industry has put forward the demand for flexibility and high precision for robot assembly.Vision-guided robotic arm assembly is an important way to achieve flexible assembly in small batches and multiple batches of 3C manufacturing.In order to meet the application requirements of high-precision flexible assembly in the 3C manufacturing industry,For several problems in high-precision vision-guided flexible assembly of small screws,facing the application requirements of highprecision flexible assembly in the 3C manufacturing industry,this thesis proposes a new dualcamera hand-eye calibration method,a depth-constrained-based screw hole edge feature extraction method,a 3D point cloud screw hole location and a new method to solve hole pose and screw pose,and develops a visual guidance system for flexible assembly of mobile phone screws,and verify the function of the developed system through experiments.The main contents of this thesis are summarized as follows:(1)Through in-depth research on various flexible assembly visual guidance technologies at home and abroad,we understand the research progress of robot screw tightening methods,analyze the key problems in the application scene in terms of accuracy and flexibility,and establish the research content and scheme of this thesis.(2)Aiming at the problem that the existing hand-eye calibration algorithm is limited by the influence of the absolute positioning accuracy of the manipulator,resulting in poor calibration accuracy,this thesis proposes a dual-camera hand-eye calibration method,that is,the cooperative target is used as position feedback to replace the manipulator.The terminal pose feedback can effectively improve the accuracy of hand-eye calibration.(3)Aiming at the problem of screw hole feature extraction in 3D point cloud,using the two-dimensional image and depth map information of structured light three-dimensional imaging process,this thesis proposes a depth-constrained screw hole edge feature extraction algorithm and 3D point cloud screw hole pose calculation the method can effectively solve the problem of 3D point cloud screw hole recognition and pose acquisition.At the same time,a motion camera-based screw pose calculation method is proposed to determine the screw end pose in the flexible assembly of mobile phone screws,and plan the movement path of the robotic arm for obstacle avoidance and multi-hole assembly tasks.(4)On the basis of the above algorithm,this thesis develops a set of robot vision guidance system for flexible assembly of mobile phone screws,and programs the key algorithms and related operation functions of flexible assembly vision guidance;The proposed algorithm and system are effective.