Recognition And Positioning Of Easily Deformed Materials Based On Stereo Vision

Product assembly is the core process of intelligent manufacturing.China has many automatic assembly lines in 3C,automobile and other industries,but the assembly in many industries is still in the stage of manual assembly,and manual assembly has low efficiency and unstable assembly quality,so the research on automatic and intelligent assembly technology is the key to solve the assembly problem.At present,there are many researches and applications on the assembly of rigid materials,but there are few researches on the assembly of easily deformed materials.This paper mainly studies the identification,positioning and assembly of materials that are prone to deformation.The difficult problem of easily deformed material assembly is that the shape is easily changed in the process of assembly.In this paper,the assembly of easily deformed wire is taken as the research object.Secondly,aiming at the problems existing in the assembling process of the easily deformed materials,a method of attitude identification and positioning of the easily deformed wire elements based on stereo vision is proposed,and the model of assembling the easily deformed wire is also studied.The main research contents of this paper are as follows:1.This paper introduces the research background and significance of this topic,analyzes the current situation at home and abroad,and summarizes the existing research status of 2D and 3D visual guidance technology,as well as the development process of 3D technology.2.Based on the analysis of the assembly process of the easily deformed wire elements(easily deformed materials)in the production process of micro motor,the key technical problems to be solved are visual identification,positioning and assembly.Firstly,the assembly requirements of the easily deformed wire in the process are analyzed,and according to the characteristics of the easily deformed wire,the hardware equipment suitable for the assembly process is determined.3.The attitude identification and positioning method of the easily deformed wire.Firstly.the point cloud data is collected by industrial camera,and the point cloud data is preprocessed.Then,the shape of the easily deformed wire is identified by using the probabilistic statistical method.Finally,by predicting the shape type of the wire,the point coordinates of the wire are calculated and the posture of the wire is obtained.4.In order to solve the problems of narrow assembly space and high assembly accuracy of easily deformed wires,a four-point hand-eye calibration algorithm based on error distribution estimation was proposed to reduce the error of coordinate transformation between robot coordinate axis and 3d camera coordinate axis.In order to solve the assembling problem of easily deformed wires,a model of assembling easily deformed wires based on state machine is proposed.Finally,in the experimental part,the method proposed in this paper is experimentally verified.The experimental results show that the recognition rate of this method is 98.4%,which can solve the problem of attitude recognition and positioning of the easily deformed wires.The average error of four-point hand-eye calibration position based on error distribution estimation is within 0.45 mm.