機械手臂的重要性隨著智慧製造的發展而與日俱增，尤其在人力成本逐漸上升之際，機械手臂更已成為維持競爭力的主要關鍵。隨著感測器技術的湧現，產業越來越追求機械手臂的彈性化，因此人們不斷鑽研以結構光掃描器與CCD探求物體在環境中的數據，並將之與機械手臂結合，使機械手臂達到彈性化的應用。現階段結構光掃描的技術是利用三維點資料計算物件的夾取位置，但此法容易因掃描視角而錯失許多更穩固的夾取選擇。為克服此問題，本研究開發以零件的CAD模型進行幾何分析的方法，讓機械手臂有更多夾取選擇，進而篩選出更穩固的夾取位置。
本研究主要針對CAD模型的幾何資訊，利用射線干涉法進行夾取點的分析，並增設夾爪檢查機制，以避免夾爪與零件產生碰撞，並利用三維點資料判別物體在工作台上的位置以及姿態，進而以特定條件計算出適當的夾取點，最後以機械手臂實行零件實物之夾取。
本論文除詳述以CAD模型分析夾取點和三維點資料處理的演算法外，也利用多個不同幾何特性的零件來驗證本分析系統的實用性。

The presence of robot arm has been turning into the core of remaining competitiveness as intelligent manufacturing has been developed, especially while labor cost is ascending. And as the technology of sensors surfaces, industries commence being in pursuit of flexibility of robot arm application, thus experts started delving into structured light scanner and CCD to try to acquire data in environment, and combine it with robot arm to attain the maximum of the flexibility in industry automation.
Robot grasping nowadays mostly utilizes only 3D point data, which is from structured light scan to evaluate grasping points. However, the result of this method thoroughly relies on the camera viewing angle, therefore, it will just furnish a robot arm with narrow grasping options. To deal with this problem, this research developed a system to analyze grasping points via 3D CAD model with intact geometric information in order to provide more various options for robot arm, on top of it, robot arm will be able to designate a most stable grasping points.
This research primarily utilizes geometric information from 3D CAD model to analyze grasping points by ray intersection. And in order to avert the interference between gripper and object, this research builds up a specific checking procedure. Moreover, it also sieves out inappropriate grasping points from all options due to present pose, and pick out the most appropriate ones via certain conditions eventually. At last, it will command robot arm to grasp the object at exact position.
Besides expatiating on PCL algorithms and 3D CAD analysis, this thesis also demonstrates the capability of the system via couple of objects with various geometries.

[1]R. B. Rusu and S.Cousins (2011), “3d is here: Point cloud library (pcl) ,” in Robotics and automation(ICRA), 2011 IEEE International Conference, May 09-13, Shanghai, China.
[2]M. Alexa and A. Adamson (2004), “On normals and projection operators for surfaces defined by point sets,” in Proceedings of the First Eurographics conference, June 02-04, Switzerland.
[3]E. Whal, U.Hillenbrand, and G. Hirzinger (2003), “Surflet-pair-relation histograms: a statistical 3D-shape representation for rapid classification,” in 3-D Digital Imagimg and Modeling, Fourth International Conference, October 06-10, Banff, Alberta, Canada.
[4]R. B. Rusu, N. Blodow, and M. Beetz (2009), “Fast point feature histograms (FPFH) for 3D registration, ” in Robotics and Automation, IEEE International Conference, May 12-17, Kobe, Japan.
[5]I. Sipiran and B. Bustos (2011), “Harris 3D: A robust extension of the Harris operator for interest point detection on 3D meshes,” Visual Computer, Vol. 27, Issue 11, pp. 963-976.
[6]R. B. Rusu, Z. C. Marton, N. Blodow and M. Beetz (2008), “Persistent point feature histograms for 3D point clouds,” in Proceedings of the 10th International Conference on Intelligent Autonomous Systems, Baden Baden, Germany.
[7]R. B. Rusu, Z. C. Marton, N. Blodow and M. Beetz (2008), “Aligning point cloud views using persistent feature histograms,” in Intelligent Robots and Systems, IEEE/RSJ International Conference, September 22-26, Nice, France.
[8]R. B. Rusu, G. Bradski, R. Thibaux, and J. Hsu (2010), “Fast 3d recognition and pose using the viewpoint feature histogram,” in Intelligent Robots and Systems, IEEE/RSJ International Conference, October 18-22, Taipei, Taiwan.
[9]a. Aldoma, M. Vincze, N. Blodow, D. Gossow, S. Gedikli, R. B. Rusu (2011), “CAD-model recognition and 6DOF pose estimation using 3D cues,” in Computer Vision Workshops, IEEE International Conference, November 06-13, Barcelona, Spain.
[10]A. Aldoma, F. Tombari, R. B. Rusu, and M. Vincze (2012), “OUR-CVFH-Oriented, Unique and Rapeatable Clustered Viewpoint Feature Histogram for Object Recognition and 6DOF Pose Estimation,” in Joint 34th DAGM and 36th OAGM Symposium, August 28-31, Graz, Austria.
[11]P. J. Besl and N. D. Mckay (1992), “ Method for registration of 3-D shapes,” IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol.14, Issue 2, pp. 239-256.
[12]S. Rusinkiewicz and M. Levoy (2001), “Efficient variants of the ICP algorithm,” in Proceedings Third International Conference on 3-D Digital Imaging and Modeling, May 28-June 01, Quebec, Canada.
[13]蔡仕晟，「擬人機械手臂基於影像之適應性抓取設計」（2011），碩士論文，交通大學電控工程研究所，台北市。
[14]Javier Romero,「From Human to Robot Grasping」（2011）, Doctoral Thesis, School of Computer Science and Communication, Stockholm, Sweden.
[15]Lawson L. S. Wong，「Robotic Grasping on the Stanford Artificial Intelligence Robot」（2008），Department of Computer Science of Stanford Univeristy, USA.
[16]辜敬忠，「視覺引導四軸機械手臂之製作與物件抓取控制」（2011），碩士論文，東南科技大學機電整合研究所，台北市。
[17]韓致偉，「3D視覺伺服器系統整合機械手臂於物體追蹤取放之研究」（2006），碩士論文，華梵大學機電工程學系，台北市。
[18]蔡銘富，「植基於立體視覺之三為物件辨識與追蹤」（2009），碩士論文，台北科技大學自動化科技研究所，台北市。
[19]黃瑞逸，「以影像視覺為基礎的2D氣壓手臂系統之追蹤控制」（2012），碩士論文，台灣科技大學機械工程系，台北市。
[20]陳志明，「使用影像伺服之機械手臂抓取系統開發研究」（2005），碩士論文，長庚大學機械工程研究所，桃園市。
[21]林姿吟，「基於立體視覺的3D影像定位」（2011），碩士論文，中原大學機械工程研究所，桃園市。
[22]陳柏均，「基於3D物件辨識與對位之機械手臂夾取系統」（2016），碩士論文，台北科技大學自動化科技研究所，台北市。
[23]譚孟桓，「機械視覺應用於四軸機械手臂之定向夾取」（2014），碩士論文，龍華科技大學機械工程研究所，桃園市。
[24]陳希哲，「應用運動回復結構技術建立機器手臂夾取點與三維物體模型」（2013）碩士論文，臺灣師範大學機電科技學系，台北市。
[25]D. C. Dupuis,「Two-fingered Grasp Planning for Randomized Bin-picking」（2009）, Doctoral Thesis, University of British Columbia, Vancouver, USA.
[26]K. Hsiao, S. Chitta, M. Ciocarlie, E. G. Jones (2010), “Contact-reactive grasping of objects with partial shape information,” in IEEE/RSJ International Conference on Intelligent Robots and Systems, Taipei, Taiwan.
[27]http://pointclouds.org/documentation/
[28]https://community.ptc.com/t5/Customization/bd-p/customization
[29]Pro/TOOLKIT User’s Guide (2009), Parametric Technology Corporation, Needham, MA, USA.
[30]王文波（2010），Pro/E Wildfire 4.0 二次開發實例解析，清華大學出版社，北京。