CASIA OpenIR  > 毕业生  > 博士学位论文
基于显微视觉的微零件位姿检测与装配研究
Alternative TitleResearch on Pose Detection of Micro-parts and Micro-assembly Base on Microscope Vision
张娟
Subtype工学博士
Thesis Advisor徐德
2013-05-22
Degree Grantor中国科学院大学
Place of Conferral中国科学院自动化研究所
Degree Discipline控制理论与控制工程
Keyword特征提取 趋近与抓取 位姿检测 自动变倍 运动控制 显微视觉 微装配 Feature Extraction Approaching And Clamping Pose Detection Automatic Zoom Movement Control Microscope Vision Micro-assembly
Abstract随着微机电系统的发展,对微小零件装配提出了更高的要求。显微视觉引导下的微装配机器人为微小零件三维空间装配提供了有效解决方案。本论文针对显微视觉微装配关键技术展开研究。本论文的主要研究工作有: 1)针对夹持器趋近抓取问题,提出基于两路显微视觉的夹持器趋近与抓取控制策略。实现了微零件任意初始位姿下,夹持器向微零件的快速稳定趋近与抓取。基于夹持器和微零件的成像特征分析,提出实时有效的显微图像特征提取算法,实现了对夹持器与微零件空间相对位置的估计。 2)由于微零件薄且脆,在微零件与夹持器姿态未对准的情况下,夹持器趋近过程中可能与微零件发生触碰造成微零件的损坏。针对该问题,在夹持器的运动控制中引入微零件的受力信息,提出视触觉混合的夹持器运动控制策略,在实现夹持器快速稳定趋近的同时,保证了微零件受力在安全范围之内。 3)由于显微视觉系统景深短,不易构成传统立体视觉,微零件三维空间位姿不易获得。基于显微视觉成像特点分析,提出了基于三路显微视觉的粗精结合的微零件在线检测与对准策略,实现了mm级复杂结构微零件空间位姿的高精度检测与对准。微零件相对姿态检测误差小于0.5度,相对位置检测误差小于5µm。 4)针对显微视觉系统高放大倍数与大视场相互矛盾的问题,提出显微视觉系统自动变倍算法,基于微零件图像特征区域跟踪以及显微视觉系统空间位置调整,实现了装配过程中对显微视觉系统放大倍数的自动调整。 5)针对微装配工艺流程复杂,自动化程度不高的问题,提出了基于三路显微视觉的微零件自动对准控制策略,实现了复杂结构微零件三维空间位姿的快速有效对准,该控制策略基于图像雅可比矩阵的视觉伺服控制方式。推导了多路显微视觉伺服控制的图像雅可比矩阵,并通过在线标定技术实现了对图像雅可比矩阵的标定。 此外,针对自动聚焦技术,在线标定技术展开研究。研究柱状微零件边缘区域自动聚焦问题,提出了粗精结合的自动聚焦策略;提出了显微视觉系统比例系数在线标定方法,实现了比例系数的在线标定。
Other AbstractWith the development of micro-electro-mechanical system, there is a higher requirement for small parts assembly. Micro-assembly robot in the guide of microscope vision provides an effective solution for the three-dimensional space assembly of small parts. The thesis focuses on the key technologies of micro-assembly in the guide of microscope vision. The main work is described as follows. 1) For gripper approaching and clamping, a control strategy of gripper based on two microscope vision systems is proposed. The gripper approaching and clamping the micro-part is fast and stable in an arbitrary initial pose of the micro-part. Based on imaging feature analysis of the gripper and the micro-part, an effective image feature extraction algorithm is proposed to estimate the relative position of the gripper and the micro-part in real-time. 2) As the micro-part is thin and brittle, in the case of the gripper is not parallel to the micro-part, the gripper may contact and damage the micro-part in approaching process. To solve this problem, the force of the micro-part is introduced into the gripper movement control process, a control strategy based on visual and force information is proposed. The gripper approaching to the micro-part is fast and stable, and the force of the micro-part is in a safe range. 3) As microscope vision has the feature of short depth of view field, it is not easy to obtain the micro-part three-dimensional space position and orientation through traditional stereo vision method. Based on microscope vision feature analysis, a pose detection and alignment strategy combination of coarse and fine process based on three microscope vision systems is proposed. The relative poses of mm sized complex structure micro-parts are detected and adjusted in high accuracy. The detection error of relative pose is less than 0.5 degree, and the detection error of relative position is less than 5µm. 4) High magnification and large field of view of microscope vision are conflicting. To solve this question, an automatically zoom algorithm is proposed. Based on image feature tracking and space position adjustment of microscope vision system, the magnification of the microscope vision is adjusted automatically to achieve the assembly process. 5) The micro-assembly process is always complex, and the degree of automation is always low. An automatic alignment control strategy based on three-way microscope vision systems is proposed to achieve complex ...
shelfnumXWLW1860
Other Identifier201018014628020
Language中文
Document Type学位论文
Identifierhttp://ir.ia.ac.cn/handle/173211/6506
Collection毕业生_博士学位论文
Recommended Citation
GB/T 7714
张娟. 基于显微视觉的微零件位姿检测与装配研究[D]. 中国科学院自动化研究所. 中国科学院大学,2013.
Files in This Item:
File Name/Size DocType Version Access License
CASIA_20101801462802(13590KB) 暂不开放CC BY-NC-SAApplication Full Text
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[张娟]'s Articles
Baidu academic
Similar articles in Baidu academic
[张娟]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[张娟]'s Articles
Terms of Use
No data!
Social Bookmark/Share
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.