本文選題：視覺 + 伺服機器人�。� 參考：《南京理工大學(xué)》2017年碩士論文

【摘要】：本文針對現(xiàn)階段工業(yè)生產(chǎn)過程中示教型機器人對環(huán)境適應(yīng)能力有限問題,提出加入視覺傳感器,為機器人裝上"眼睛和大腦",結(jié)合機器人視覺伺服控制研究與應(yīng)用現(xiàn)狀,以視覺伺服機器人抓取連續(xù)運動傳送帶上條煙為研究對象,完成了對條煙的分類識別與定位抓取。為實現(xiàn)這一目的,主要進行了以下研究:(1)動態(tài)條煙抓取系統(tǒng)總體方案設(shè)計。為滿足項目需求,確定系統(tǒng)總體工作流程。根據(jù)精度要求和產(chǎn)品性價比,對方案中涉及到的軟硬件進行選型,完成實驗平臺搭建。(2)條煙分類識別與定位研究。通過對條煙圖像進行預(yù)處理、邊緣檢測與跟蹤、創(chuàng)建模板、尋找模板等研究,獲取條煙在圖像上的精確位置。通過基于支持向量機的分類方法,實現(xiàn)條煙的準確分類。(3)基于Halcon的系統(tǒng)標定與抓取研究。進行相機標定、手眼標定,并對手眼標定進行校正,實現(xiàn)坐標系統(tǒng)一。對機械手末端進行路徑軌跡規(guī)劃,計算條煙在運動傳送帶上的抓取位置,實現(xiàn)對條煙的準確抓取。(4)系統(tǒng)外圍電路整合設(shè)計。運用分層思路實現(xiàn)外圍電路系統(tǒng)總體設(shè)計,對不同模塊完成軟硬件設(shè)計,并以光纖觸發(fā)光源頻閃為例介紹外圍電路工作流程。(5)動態(tài)條煙抓取系統(tǒng)設(shè)計與開發(fā)。以Visual Studio為開發(fā)平臺,結(jié)合Windows的MFC功能與消息映射機制,對系統(tǒng)進行基于對話框的設(shè)計。分模塊實現(xiàn)上位機與機器人控制器串口通信、相機SDK開發(fā)功能,同時增加異常處理機制,最終完成條煙分類檢測與定位抓取系統(tǒng)的設(shè)計與開發(fā)。
[Abstract]:In this paper, aiming at the problem of limited adaptability of teaching robot to environment in the present industrial production process, this paper puts forward adding vision sensor to install "eyes and brain" for robot, combining with the research and application status of robot visual servo control.In this paper, a visual servo robot grabs a cigarette on a continuous moving conveyer belt as the research object, and realizes the classification recognition and location grabbing of the strip smoke.In order to achieve this goal, the following research is mainly carried out: 1) the overall scheme design of dynamic cigarette grab system.To meet the requirements of the project, determine the overall system workflow.According to the requirement of precision and the ratio of product performance to price, the software and hardware involved in the scheme were selected, and the experimental platform was built.In order to obtain the exact position of cigarette on the image, the cigarette image is preprocessed, edge detection and tracking, template creation, template searching and so on.The classification method based on support vector machine (SVM) is used to realize the accurate classification of stripes. (3) the system calibration and grabbing based on Halcon are studied.Camera calibration, hand-eye calibration, and hand-eye calibration are carried out to achieve the unification of coordinate system.The path path planning is carried out at the end of the manipulator, and the grab position of the smoke on the moving conveyor belt is calculated, and the integrated design of the peripheral circuit of the system is realized.The overall design of peripheral circuit system is realized by using delamination, and the hardware and software of different modules are designed. The design and development of dynamic cigarette grab system are introduced with the example of fiber trigger light source stroboscopic.Taking Visual Studio as the development platform, combining the MFC function of Windows and message mapping mechanism, the system is designed based on dialog box.The module realizes the serial communication between the upper computer and the robot controller, the SDK development function of the camera, and the exception handling mechanism. Finally, the design and development of the smoke classification detection and location grab system are completed.
【學(xué)位授予單位】：南京理工大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2017
【分類號】：TP242

【參考文獻】

相關(guān)期刊論文 前9條

1 李長俊;張曉陽;;鍋仔片表面指紋檢測方法[J];計算機測量與控制;2014年08期

2 柳賀;李勛;劉蕾;;工業(yè)機器人可靠性設(shè)計與測試研究[J];中國新技術(shù)新產(chǎn)品;2014年14期

3 崔帥鋒;;基于HALCON軟件的單攝像機標定方法研究[J];中國科技信息;2014年Z1期

4 鄧彩霞;王貴彬;楊鑫蕊;;改進的抗噪形態(tài)學(xué)邊緣檢測算法[J];數(shù)據(jù)采集與處理;2013年06期

5 姜富修;姚淮林;;淺析智能化開關(guān)設(shè)備[J];機械制造與自動化;2013年05期

6 倪受東;丁德健;張敏;;視覺功能六自由度工業(yè)機器人的研制[J];制造業(yè)自動化;2012年24期

7 潘麗霞;;論工業(yè)機器人的發(fā)展與應(yīng)用[J];山西科技;2010年03期

8 謝俊峰;龔健雅;江萬壽;;一種改進的恒星相機在軌檢校方法[J];測繪科學(xué);2009年02期

9 高紅波;王衛(wèi)星;;一種二值圖像連通區(qū)域標記的新算法[J];計算機應(yīng)用;2007年11期

，

本文編號：1760790