臺灣博碩士論文加值系統

本論文旨在開發EtherCAT工業乙太網路結合雙目視覺與深度學習之六軸機械手臂。系統可區分為眼、腦及手臂三部分，其中雙目視覺為系統的眼睛，負責擷取外界場景；深度學習是系統的大腦，負責場景中目標識別與定位；而可移動機械臂系統由移動滑台和固定在其上的EtherCAT六自由度機械手臂組成，負責抓取目標。在硬體部分，以圓管式臂桿及中空式關節結構，進行可移動六自由度仿人機械手臂機構設計，並分別以直流馬達、中空式諧和減速機與雙目立體視覺辨識系統，完成機械手臂的驅動設計及物件辨識。軟體部分，以工業電腦為核心具有EtherCAT通訊功能，結合LabVIEW圖控軟體進行人機界面與控制程式的撰寫，完成整個軟體系統的設計。因此，研製具有雙目視覺與深度學習六軸機械手臂，並結合模組化關節控制進行手臂的設計，使手臂不僅能重複執行固定的任務，且可依據物件的顏色、大小及形狀執行不同的工作任務。

This thesis develops a six-axis manipulator that combines the industrial communication of EtherCAT with binocular vision and deep learning. The system can be divided into three parts: eyes, brain and arm. Binocular vision is the eyes of this system, responsible for capturing the external scene; deep learning is the brain of this system, responsible for the object recognition and localization in the scene; and the mobile manipulator system consists of a sliding platform and an EtherCAT six-DOF manipulator fixed on it, which is responsible for grabbing localization. In the hardware part, the mobile six-DOF humanoid manipulator mechanism is equipped with DC motors, hollow harmonic drives and a binocular stereo vision recognition system, which realizes the mechanism design of round tubes and hollow joints to complete the drive design and object recognition. The software part, the core adopts an industrial computer with EtherCAT communication function. Write the human machine interface and control program to complete the design of the entire software system by using LabVIEW software. Therefore, the development of the six-axis manipulator with binocular vision and deep learning, combined with modular joints control, allows the manipulator to not only perform repeated tasks, but also perform different tasks depending on the color, size and shape of the object.

摘要 I
Abstract II
謝誌 III
目錄 IV
表目錄 VII
圖目錄 VIII
第一章 緒論 1
1.1 研究背景與動機 1
1.2文獻回顧 2
1.3論文架構 8
第二章 系統架構與實驗設備 9
2.1 具有雙目視覺與深度學習六軸機械手臂設計與建立 9
2. 2 電控系統設計 17
2.3 實驗系統設備 20
2.3.1 馬達 20
2.3.2 馬達驅動器 20
2.3.3 單軸機器人 21
2.3.4減速機 22
2.3.5伺服驅動系統 23
2.3.6夾爪 23
2.3.6 EtherCAT Master Card 24
2.3.7 3D視覺感測模組 24
2.3.8 PC-Based控制系統與使用環境 25
第三章 深度學習 26
3.1 卷積神經網路 26
3.1.1 卷積層 27
3.1.2 池化層 28
3.2 Regions with CNN 29
3.3 Fast Regions with CNN 30
第四章 實驗結果 31
4.1 實作一 31
4.2 實作二 38
第五章 結論與未來展望 42
5.1 結論 42
5.2 未來展望 42
參考文獻 44

[1] "Unimate," Robotic Industries Association, 2018. [Online]. Available: https://www.robotics.org/joseph-engelberger/unimate.cfm.
[2] S. Y. Nof, Handbook of Industrial Robotics, NY: John Wiley & Sons, 1999.
[3] M. H. Ang, L. Wei, S. Y. Lim, "An Industrial Application of Control of Dynamic Behavior of Robots - A Walk-Through Programmed Welding Robot," in IEEE International Conference on Robotics and Automation, San Francisco, USA, 2000.
[4] 劉咏龍，適應性視覺伺服控制之機械手臂應用，碩士論文，國立中央大學電機工程研究所，桃園，2003。
[5] M. Galicki, "Control-Based Solution to Inverse Kinematics for Mobile Manipulators Using Penalty Functions," Journal of Intelligent and Robotic Systems, vol. 42, no. 3, p. 213–238, 2005.
[6] H. P. Huang, Y. T. Chen, R. J. Wang, M. K. Chi, "Development of the Multi-Axis Control Platform for Robot Arm," in Proceedings of SICE Annual Conference, Taipei, Taiwan, 2010.
[7] 鍾玉琢、喬秉新、李樹青，機器人視覺技術，北京：國防工業出版社，1994。
[8] 魏士勛，多攝影機重疊區域人流修正技術之研究，碩士論文，大同大學資訊工程學系，臺北，2010。
[9] M. C. Lu, W. Y. Wang, C. Y. Chu, "Image-Based Distance and Area Measuring Systems," IEEE Sensors Journal, vol. 6, no. 2, pp. 495-503, 2006.
[10] A. Nakashima, Y. Sugiyama, Y. Hayakawa, "Paddle Juggling of one Ball by Robot Manipulator with Visual Servo," in International Conference on Control, Automation, Robotics and Vision, Singapore, 2006.
[11] V. Lippiello, B. Siciliano, L. Villani, "Position-Based Visual Servoing in Industrial Multirobot Cells Using a Hybrid Camera Configuration," IEEE Transactions on Robotics, vol. 23, no. 1, pp. 73-86, 2007.
[12] J. K. Oh, C. H. Lee, "Development of a Stereo Vision System for Industrial Robots," in International Conference on Control, Automation and Systems, Seoul, South Korea, 2007.
[13] "EtherCAT Technology Group," EtherCAT技術協會, 2018. [Online]. Available: https://www.ethercat.org/default.htm.
[14] M. Rostan, "High Speed Industrial Ethernet for Semiconductor Equipment," SEMI Technology Symposium: Innovations in Semiconductor Manufacturing, San Francisco, CA, 2004.
[15] 向乾亮，實時乙太網EtherCAT系統設計及在電力系統中的應用，碩士論文，華北電力大學，北京，2008。
[16] 葛世偉、廖慶財，EtherCAT 在防災系統之應用，工程科技與教育學刊，第12卷，第1期，第8-13頁，2015。
[17] G. E. Hinton, R. R. Salakhutdinov, "Reducing the Dimensionality of Data with Neural Networks," Science, vol. 313, pp. 504-507, 2006.
[18] D. Erhan, Y. Bengio, A. Courville, P. A. Manzagol, P. Vincent, "Why Does Unsupervised Pre-training Help Deep Learning?," Journal of Machine Learning Research, vol. 11, no. 3, pp. 625-660, 2010.
[19] A. Krizhevsky, I. Sutskever, G. E. Hinton, "ImageNet Classification with Deep Convolutional Neural Networks," in International Conference on Neural Information Processing Systems, Nevada, USA, 2012.
[20] D. Silver, A. Huang, C. J. Maddison, A. Guez, L. Sifre, G. van den Driessche, J. Schrittwieser, I. Antonoglou, V. Panneershelvam, M. Lanctot, S. Dieleman, D. Grewe, J. Nham, N. Kalchbrenner, I. Sutskever, T. Lillicrap, M. Leach, K. Kavukcuoglu, T. Graepel, "Mastering the game of Go with deep neural networks and tree search," Nature, vol. 529, pp. 484-489, 2016.
[21] A.S. Razavian, H. Azizpour, J. Sullivan, S. Carlsson, "CNN Features Off-the-Shelf: An Astounding Baseline for Recognition," in IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, Columbus, USA, 2014.
[22] "DCX系列馬達," MAXON MOTOR, 2018. [Online]. Available: https://www.maxonmotor.com.tw/maxon/view/content/Overview-Page-DC-Motors.
[23] "MAXPOS," MAXON MOTOR, 2018. [Online]. Available: https://www.maxonmotor.com.tw/maxon/view/content/Overview-Page-Controls.
[24] "單軸機器人," 全球傳動科技股份有限公司, 2018. [Online]. Available: https://www.tbimotion.com.tw/zh-TW/category/CAT-Single-Axis-Actuator.html.
[25] "SHD-2SH系列," Harmonic Drive, 2018. [Online]. Available: http://www.harmonicdrive.net/products/gear-units/simplicity-gear-units/shd-2sh.
[26] "伺服驅動系統," 台達電子工業股份有限公司, 2018. [Online]. Available: http://www.deltaww.com/Products/CategoryListT1.aspx?CID=060201&PID=14&hl=zh-TW&Name=ASDA-A2.
[27] "PCI-1203," 研華股份有限公司, 2018. [Online]. Available: https://www.advantech.tw/products/7d3c9775-8c30-4f65-83ec-755bee93b1d4/pci-1203/mod_8a5fd662-913c-4bb4-b7d4-a01cb12edcfb.
[28] "3D視覺模組," 所羅門股份有限公司, 2018. [Online]. Available: https://www.solomon.com.tw/product-category/vision-robotics/3d-vision-robotics/.
[29] "IPC-7130," 研華股份有限公司, 2018. [Online]. Available: http://www.advantech.tw/products/1-2jkemv/ipc-7130/mod_d2b390ec-5003-4cfe-b9e6-80917ffbb3a4.
[30] 斎藤康毅，譯者：吳嘉芳，Deep Learning用Python進行深度學習的基礎理論實作，臺北：歐萊禮，2017。
[31] P.Y. Simard, D. Steinkraus, J.C. Platt, "Best Practices for Convolutional Neural Networks. Applied to Visual Document Analysis," in International Conference on Document Analysis and Recognition, Edinburgh, UK, 2003.
[32] D. Scherer, A. Müller, S. Behnke, "Evaluation of Pooling Operations in Convolutional Architectures for Object Recognition," in International Conference on Artificial Neural Networks, Thessaloniki, Greece, 2010.
[33] R. Girshick, J. Donahue, T. Darrell, J. Malik, "Rich feature hierarchies for accurate object detection and semantic segmentation," in IEEE Conference on Computer Vision and Pattern Recognition, Columbus, USA, 2014.
[34] J. R. R. Uijlings, K. E. A. van de Sande, T. Gevers, A. W. M. Smeulders, "Selective Search for Object Recognition," International Journal of Computer Vision, vol. 104, no. 2, pp. 154-171, 2013.
[35] R. Girshick, "Fast R-CNN," in IEEE International Conference on Computer Vision, Santiago, Chile, 2015.
[36] 林唯修，六軸機械手臂之研製與位置控制，碩士論文，國立臺灣科技大學高分子工程系，臺北，2009。