Research On 3D Pose Reconstruction Method And Precision Of Component Pin Rotation Vision

In the vision system of the special-shaped plug-in machine,considering that the electronic components tend to be diversified and refined,and the PCB integration is getting higher and higher,the precision requirements of the plug-in machine and other component assembly equipment are continuously improved.In view of the plug-in machine vision system requiring high accuracy,high real-time performance,narrow system structure space,and the current incompleteness of the rotation vision pin position and attitude measurement theory in accuracy analysis,in order to meet the requirements of more and more precision and speed Based on the rotating stereo vision system,this paper proposes two methods of reconfiguration of stitch poses,and performs error analysis on the main system parameters and structural parameters based on the built mathematical model to solve the problems related to the accuracy of automatic insertion of special-shaped components.Considering the problem that the current least squares fitting conventional fitting algorithm cannot meet the accuracy requirements in the face of highprecision application environment,this paper improves the existing least squares fitting method,and proposes a weighted fitting method and relative position constraints The stitch pose fitting method of method,multi-point mapping average method,and verify the proposed pose estimation optimization algorithm through simulation and experiment.This paper first establishes a system parameter model from the image coordinate system to the world coordinate system based on the three-dimensional coordinate vision system at the end of the stitch,and derives the algebraic reconstruction method of stitch pose reconstruction based on the principle of rotating vision.Based on the algebraic reconstruction method,a trigonometric reconstruction method of equivalent binocular vision is proposed.On the basis of the stitch pose estimation method based on least square fitting,the weighted reconstruction method,the relative position constraint method(including the circle constraint and the invariant angle constraint)and the high-precision position of the multi-point mapping average method are further proposed.Pose estimation algorithm.Based on the stitch reconstruction model,a quantitative analysis method for the influence of the accuracy of key rotational vision components and key parameters such as rotation axis accuracy,pixel resolution,lens focal length,rotation radius and stitch height on the accuracy of stitch pose estimation is derived through theoretical analysis,and The error analysis model established by simulation verification is obtained,and the relationship between some important parameters of the vision system and the estimation accuracy of stitch pose is obtained.By analyzing the relationship between the imaging uncertainty and the stitch pose,a threedimensional map of the image point error and the stitch pose estimation error is obtained.With the aid of the triangulation analysis method of equivalent binocular vision,it is found that the stitch pose measurement at certain positions Reasons and rules of abnormal accuracy;based on the problem of different positioning accuracy of stitch positions at different positions,the rationality of using weighted reconstruction is verified,and the optimal rotation vision system based on two image reconstructions is obtained The angle combination of initial angle and rotation angle.Using the combined analysis method,through simulation analysis,the maximum error between the fitting method based on circle constraint and the traditional method based on least square fitting in the single-pin multi-angle reconstruction under different parameter configurations is compared,and the number of imaging points and resolution are obtained.,Noise level,etc.on the accuracy of the law.On the basis of verifying the effectiveness of the circle constraint for accuracy improvement,the fitting method of the relative position constraint is composed of the invariant angle constraint and the circle constraint through simulation analysis,and the multi-point mapping averaging method based on the principle of smoothing noise reduction and the traditional minimum The comparison of the square fitting methods proves that the method proposed in this paper can effectively improve the accuracy of stitch pose estimation.Finally,an experimental platform for rotating vision positioning accuracy is built,and the simulation conclusions of this paper and the effectiveness of the proposed stitch pose estimation algorithm are verified based on actual working conditions.