Research on Machine Vision Based Precise Assembly for Miniature Parts
|School||Dalian University of Technology|
|Course||Mechanical and Electronic Engineering|
|Keywords||The micro small parts assembly Machine Vision Precision assembly systems Automated assembly|
Manufacturing of precision miniature devices and systems products, the size of the required assembly parts generally include several tenths of millimeters to a dozen millimeters, high assembly precision requirements. Assembly operations are generally under a microscope by a skilled operator manually, production efficiency and lower yield, poor product consistency, higher skill requirements and operating personnel. With the continuous improvement of product performance, quality and stability requirements and production needs, the manual assembly is difficult to adapt to. Assembly system based on visual characteristics of its flexible assembly to improve the assembly accuracy, and ensure the stability of product quality and increase productivity, are showing great advantages. In micro parts visual assembly system, the visual part of the main task is to identify small objects, measuring and positioning, i.e. under a certain accuracy requirements to obtain the distance to the target object, the shape, position, attitude, and other information. The article is intended for sophisticated, complex micro devices and systems products, automation, batch manufacturing, as well as to improve the quality and consistency of product, the development of micro and small parts assembly system based on machine vision. Developed a system to solve the key issues, including the size of the parts to be assembled difference, shape and characteristics of diverse surface quality difference, while ensuring the typical problems of the assembled position accuracy and geometric tolerance in a small product assembly . The micro parts automated assembly systems, including four functional subsystems: human-computer interaction, computer control subsystem, the visual subsystem clamping operation subsystem, rotary table subsystem. Visual subsystem and the preparation of a complete system control program. The visual subsystem hardware consists of a three-axis precision motion platforms, industrial digital camera and telecentric lens. First, the visual system acquisition target part image, and the image preprocessing, including smoothing filter, enhanced binarization, the mathematical morphological changes, image stitching; then extracted by the edge of the closed contour search, template matching method on the target parts identify, using the smallest enclosing rectangle France and has sub-pixel accuracy of the region centroid and the spindle of the positioning method of access to parts of the bit posture; last through the space coordinates obtained operating mechanical hand of the motion information, control manipulator completion of five different tiny parts precision assembly. The experiments show that the system developed to meet the assembly precision, improved product consistency and production efficiency, reduce the requirements for operating personnel and work intensity.