發(fā)布時間：2018-06-02 10:38

  本文選題：球閥零件裝配 + 機(jī)器視覺��； 參考：《太原科技大學(xué)》2014年碩士論文

【摘要】：球閥是生產(chǎn)生活中不可或缺的一類零件，其應(yīng)用廣、需求大，在國民經(jīng)濟(jì)中占有重要地位，裝配是球閥生產(chǎn)的后置工序，在其生產(chǎn)制造中占有重要的地位。目前，國內(nèi)企業(yè)在生產(chǎn)球閥時，仍由熟練工人運用傳統(tǒng)人眼識別的方式對其進(jìn)行裝配。這種方法容易受到工人責(zé)任心、技術(shù)水平以及疲勞程度等不穩(wěn)定因素的影響，其已不能滿足現(xiàn)代生產(chǎn)制造自動化、智能化、高精度高效率的發(fā)展要求。為此探求良好技術(shù)開發(fā)平臺下的自動裝配技術(shù)成為必然。 自機(jī)器視覺技術(shù)出現(xiàn)以來，其就被不斷地應(yīng)用于各類場合，并且取得了快速的發(fā)展。在工業(yè)生產(chǎn)中使用機(jī)器視覺系統(tǒng)即為工業(yè)視覺系統(tǒng)，按照其使用功能的不同，一般可分為檢測、定位和裝配三類。本文在球閥裝配中采用機(jī)器視覺技術(shù)，開發(fā)出基于視覺的機(jī)器人球閥智能裝配系統(tǒng)。系統(tǒng)以CCD相機(jī)對零件圖像的采集作為智能裝配的“感知”階段；編寫的系統(tǒng)控制處理程序作為智能裝配的“決策”階段；機(jī)械手爪抓取等作為智能裝配的“執(zhí)行”階段。 本文對基于視覺的機(jī)器人球閥智能裝配系統(tǒng)進(jìn)行了研究，整個系統(tǒng)由人機(jī)交互控制、裝配工作臺、視覺以及機(jī)器人夾持作業(yè)等四個主要單元組成。設(shè)計了由2臺CCD相機(jī)及工業(yè)鏡頭組成的視覺子單元、由氣動柔性通用機(jī)械手爪組成的夾持作業(yè)子單元等；同時編寫了圖像采集處理以及相機(jī)標(biāo)定等系統(tǒng)程序。首先，視覺子單元采集目標(biāo)零件圖像，，并對圖像進(jìn)行小波去噪、增強(qiáng)、二值化、數(shù)學(xué)形態(tài)計算等在內(nèi)的預(yù)處理；然后通過邊緣檢測、模板匹配等方法對目標(biāo)零件進(jìn)行識別定位；最后經(jīng)過LabVIEW處理程序得出零件在機(jī)器人局部坐標(biāo)系即標(biāo)定坐標(biāo)系中的位姿信息，進(jìn)而控制柔性機(jī)械手爪完成抓取、裝配作業(yè)。 通過現(xiàn)場實驗，驗證了所研制的視覺裝配系統(tǒng)不僅能夠滿足零件裝配精度要求，而且提高了產(chǎn)品的合格率和生產(chǎn)率，降低了對操作人員的要求和工作強(qiáng)度。
[Abstract]:Ball valve is an indispensable part in production and life, it is widely used, it has a large demand, it occupies an important position in the national economy, assembly is the post-process of ball valve production, and it occupies an important position in its production and manufacture. At present, in the production of ball valves, domestic enterprises are still assembled by skilled workers using traditional human eye recognition. This method is easy to be influenced by the unstable factors such as worker's sense of responsibility, technical level and fatigue degree. It can not meet the requirements of modern production automation, intelligence, high precision and high efficiency. Therefore, it is inevitable to explore the automatic assembly technology under the good technology development platform. Since the emergence of machine vision technology, it has been continuously applied in various situations, and has made rapid development. The use of machine vision system in industrial production is called industrial vision system. According to the different functions of machine vision system, it can be divided into three types: detection, positioning and assembly. In this paper, a robot ball valve intelligent assembly system based on vision is developed by using machine vision technology in ball valve assembly. The system takes the acquisition of part image by CCD camera as the "perception" stage of intelligent assembly, the system control processing program as the "decision" stage of intelligent assembly, and the grasp of manipulator claw as the "execution" stage of intelligent assembly. In this paper, the intelligent assembly system of robot ball valve based on vision is studied. The whole system is composed of four main units: human-computer interaction control, assembly table, vision and robot clamping operation. A visual subunit composed of two CCD cameras and industrial lens is designed, which is composed of pneumatic flexible general manipulator claw, etc. The system programs such as image acquisition and processing and camera calibration are also written. Firstly, the image of target parts is collected by visual subunit, and the image is pre-processed by wavelet denoising, enhancement, binarization and mathematical form calculation, and then the target parts are identified and located by edge detection, template matching and so on. Finally, the position and attitude information of the parts in the local coordinate system of the robot is obtained through the LabVIEW processing program, and then the flexible manipulator claw is controlled to complete the grab and assembly operation. Through field experiments, it is verified that the developed visual assembly system can not only meet the requirements of assembly accuracy, but also improve the qualified rate and productivity of the products, and reduce the requirements and working intensity of the operators.
【學(xué)位授予單位】：太原科技大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2014
【分類號】：TP242;TH134

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 王麗英;;需求帶來市場,機(jī)器視覺迎來發(fā)展的春天[J];今日電子;2008年06期

2 周秀榮;崔小虹;邢冀川;;基于LabVIEW平臺的圖像采集與處理技術(shù)研究[J];光學(xué)技術(shù);2007年S1期

3 章煒;;機(jī)器視覺技術(shù)發(fā)展及其工業(yè)應(yīng)用[J];紅外;2006年02期

4 黃桂平,李廣云,王保豐,葉聲華;單目視覺測量技術(shù)研究[J];計量學(xué)報;2004年04期

5 王曉東;程新宇;徐征;羅怡;劉超;;基于機(jī)器視覺的微小型零件精密裝配(英文)[J];納米技術(shù)與精密工程;2009年04期

6 許童羽;樸在林;陳春玲;;希爾伯特變換與小波去噪提高暫態(tài)電能質(zhì)量擾動檢測與定位精度[J];農(nóng)業(yè)工程學(xué)報;2012年19期

7 張嘉易;路超;郝永平;;微裝配系統(tǒng)結(jié)構(gòu)設(shè)計及位移臺誤差分析[J];機(jī)械設(shè)計與制造;2014年01期

8 秦益霖;孔峰;;基于LabVIEW的單目視覺灌裝桶口自動檢測與定位[J];實驗室研究與探索;2012年02期

9 查英;劉鐵根;杜東;江俊峰;;基于機(jī)器視覺的零件自動裝配系統(tǒng)[J];天津大學(xué)學(xué)報;2006年06期

10 曾歡;王浩;;圖像邊緣檢測算法的性能比較與分析[J];現(xiàn)代電子技術(shù);2006年14期

本文編號：1968456