| The accuracy of spatial geometry measurement of miniature parts determines the assembly quality and efficiency of micro-assembly system.At present,due to the advantages of continuity,flexibility and high-precision of visual measurement,it is one of the preferred measurement methods for micro-assembly system.However,the measurement dimension of monocular vision is limited,it can’t accurately measure the three-dimensional pose information of miniature parts,and in the complex assembly environment,the geometric interference of condition environment can’t guarantee the miniature parts to be in the visual imaging range.Therefore,how to improve the environmental adaptability of the visual measurement system and accurately-quickly measure the spatial pose of the miniature parts are the key to improve the assembly quality and efficiency of the micro-assembly system.In this paper,the inclinometer with fast response and strong anti-interference is introduced into the measurement system,and a combined measurement system is composed with orthogonal binocular vision.A method based on the combination of orthogonal binocular vision and inclinometer to measure the relative position and orientation of space is introduced,to realize the precise measurement and automatic assembly of trans-scale miniature parts.The basic process of imaging and the principle of camera calibration are researched,combined with the rigid transform theory construct the orthogonal binocular vision measurement model,and the three-dimensional coordinate of feature point are solved.Based on the three perpendicular theorem,the output relationship between the inclinometer and the attitude angle is derived,the position transform relation between the coordinate systems of the combined measurement system is established,and then the combined measurement model of the spatial position and attitude of micro and miniature parts is constructed.Due to the high accuracy and anti-interference ability of inclinometer,the environment adaptability and pose measurement accuracy of the combined measurement system are ensured.The interference factors in the imaging process are listed and the sources of image noise are classified.In order to solve the problem of lack the visual details in imaging,and several forms of gray transformation are deduced.Two coordinate extracting algorithms combined with image preprocessing were used to extract the coordinates of the circle center of the calibration board,to compare the accuracy of the two algorithms and provide the stable and reliable coordinate extracting algorithm for the combined measurement system.A high-precision automatic micro-assembly experiment platform is built,and the fiber array and jumper are taken as the measurement and assembly objects,to verify the accuracy and reliability of the combined measurement system.In order to ensure the accuracy of the combined measurement model,the 2D profilometer is used to measure and solve the fiber array and jumper time-sharing,and the accuracy of the spatial pose measurement model is verified by comparing with the adjustment value of the 6-DOF Platform.The source of measurement error and uncertainty is analyzed,and the corresponding uncertainty evaluation method is selected.According to the mapping relationship between input and output,the sensitivity coefficient of measurement value is deduced,and the uncertainty component is synthesized.Based on the mathematical model of combined pose measurement,the measurement uncertainty model of the system is constructed,and the uncertainty components are analyzed and solved to verify the reliability of the measurement results of the combined measurement system. |
