臺灣博碩士論文加值系統

並聯式機器人在生產線上具有十分廣泛的用途，在工業轉型自動化走向高產能高精度降低人事成本扮演十分重要的角色，目前國外已經有許多工廠廣泛應用並聯式機器人在食物封裝，零件整列等用途。並聯式機器人具有高速移動與高精度的優點，為了使其高速移動的優勢有效地應用在產線中，相匹配的視覺伺服系統也是十分重要的一環。為了達到即時追蹤工件並使並聯式機器人能夠與工件同步，本系統必須具備高速的物體辨識與追蹤系統，同時能夠精確的控制並聯式機器人的運動。
本研究為一工件追蹤與辨識的影像處理系統，可應用在高速並聯式四軸機械手臂、多軸機械手臂，等工業組裝流程上。我們設計了一“動”一“靜”兩套相結合的視覺伺服系統，希望最大程度的擴展並聯式機器人的“視覺”行程。在更早的位置、更大的視野的情況下，利用Eye-to-hand系統對工件做粗略辨識，確定吸嘴需要等在什麼位置等待工件，才能使工件能通過其夾爪上Eye-in-hand相機的視野。而後吸嘴通過高解析度的相機完成即時及精確的追&;#36394;。
針對機械手臂，本視覺整合影像演算法能夠有效率的利用視覺系統辨識工件的位置、旋轉方向與角度，並且能夠持續追蹤工件運動，進一步與機械手臂進行整合，幫助機械手臂在與物件保持一定運動以後，準確將其抓取並進行整列。傳送帶上的物件，允許其隨意位置及方向姿態，透過視覺伺服系統，機械手臂能夠準確判斷其抓取點及旋轉角度，將所有物件按照要求的姿態重新擺放。利用機械手臂，對物件做高速追&;#36394;，是本視覺整合影像處理系統一大特色，不但結合了視覺算法提供即時修正，更體現了高速並聯式四軸機械手臂本身的速度優勢。以上場景設計，皆以實際工廠情境為考量，兼具學術與產業創新應用的雙重價值。

The objective of this research is to develop conveyer object tracking and fetching system. The traditional visual servo system for parallel kinematic mechanism robot is based on Eye-to-hand method which suffers from the narrow field of view (FOV). The basic step of moving assembling is multiple objects tracking which needs broad work space and big FOV of the parallel kinematic mechanism robot. In this paper, a Hybrid Eye-to-hand and Eye-in-hand Visual Servo System for parallel kinematic mechanism robot is proposed. Via this hybrid system, the parallel kinematic mechanism robot can achieve (multiple) object tracking and fetching with priority planning we designed. The eye-to-hand visual system with low resolution camera provides the estimations of the velocities and poses of the workpieces. Also the fetching priority is planned with this system. All aforementioned information is transferred to the eye-in-hand visual servo system for further use. Once the object appeared in the FOV of eye-in-hand system in the estimated position, the high resolution camera of the system is employed to acquire the pose information of the object with high accuracy and track it frame by frame. The hybrid system is successfully demonstrated and implemented on parallel kinematic mechanism robot which not only broadens the working space of parallel kinematic mechanism robot but also achieves great efficiency and accuracy.

誌謝 i
中文摘要 ii
ABSTRACT iii
Table of Contents iv
List of Figures vi
List of Tables viii
Chapter 1 Introduction 1
1.1 Era of Robot 1
1.2 Parallel Kinematic Mechanism 6
1.3 Visual Servo System for Parallel Kinematic Mechanism Robot 7
1.4 Thesis Organization 11
Chapter 2 Image Processing 13
2.1 Otsu’s Binarization 13
2.2 Distance Transform 15
Chapter 3 Hardware Architecture 19
Chapter 4 Shortest Path 31
Chapter 5 Models of System 35
5.1 Inverse Kinematics 35
5.2 Camera Projection Model 39
5.3 Coordinate Transformation 42
5.3.1 Image Jacobian 42
5.3.2 Transformation Between Eye-Hand Systems 46
Chapter 6 Visual Servo System 47
6.1 System Description 48
6.1.1 The Eye to Hand System 48
6.1.2 The Eye in Hand System 49
6.1.3 System Flow 49
6.2 Methods 52
6.2.1 Workpiece Detection 52
6.2.2 Fetching Point Estimation 53
6.2.3 Orientation Estimation 55
6.2.4 Velocity Estimation with KF 55
6.2.5 Priority Planning 57
6.3 Error Compensation 60
6.3.1 Accuracy Definition 60
Chapter 7 Experimental Results 69
7.1 Eye-to-Hand Alone 71
7.2 Eye-in-Hand Alone 71
7.3 Hybrid System 72
Chapter 8 Conclusions and Contributions 74
Chapter 9 Future Works 75
References 76

[1]Intelligent Robotics and Automation Research(NTU-iCeiRA). Available at: www.iceira.ntu.edu.tw/. Accessed 30 April 2014.
[2]Aldebaran Robotics. Available at: www.aldebaran.com. Accessed 30 April 2014.
[3]M. King, B. Zhu, and S. Tang, “Optimal path planning,” Mobile Robots, vol. 8, no. 2, pp. 520-531, March 2001.
[4]D. Deblaise, P. Maurine, “Effective geometrical calibration of a delta parallel kinematic mechanism robot used in neurosurgery” IROS ,2005, Page(s): 1313 – 1318
[5]O. Bebek, M. Hwang, and M. Cavusoglu, “Design of a parallel kinematic mechanism robot for needle-based interventions on small animals,” IEEE/ASME Transactions on Mechatronics, vol. 18, no. 1, pp. 62–73, 2011.
[6]Y. Li, “Dynamic Analysis of a Modi&;#64257;ed DELTA Parallel kinematic mechanism robot for Cardiopulmonary Resuscitation,“ Intelligent Robots and Systems (IROS), pp. 233 – 238, 2005.
[7]M. Shoham, E. Zehavi, L. Joskowicz, E. Batkilin, and Y. Kunicher, “Bone-mounted miniature robot for surgical procedures: Concept and clinical applications,” IEEE Trans. Robot. Automat., vol. 19, no. 5, pp. 893–901, 2003.
[8]FANUC CORPORATION. Available at: www.fanuc.co.jp/en/product/index.htm. Accessed 28 April 2014.
[9]A. M. Y. Amirat, J. Pontnau, F. Artigue, “A three-dimensional measurement system for robot applications,” Journal of Intelligent and Robotic Systems, vol. 9, 1994, pp. 291-299.
[10]D. Kragic, H. Christensen, “A Framework for Visual Servoing,” International Conference on Computer Vision Systems , ICVS 2003, Graz, Austria
[11]J.M. Sebastia&;#769;n, E. Castillo, F. Roberti, R. Carelli, “One camera in hand for kinematic calibration of a parallel kinematic mechanism robot Traslosheros,” Intelligent Robots and Systems (IROS), 2010 IEEE/RSJ
[12]A. Traslosheros, J.M. Sebastia&;#769;n, E. Castillo, F. Roberti, R. Carelli, “A method for kinematic calibration of a parallel kinematic mechanism robot by using one camera in hand and a spherical object,” International Conference on Advanced Robotics, pp. 75 – 81, 2011.
[13]L. Qiu, Q. Song, J. Lei, Y. Yu, and Y. Ge, “Multi-camera-based robot visual servoing,” in Proc. IEEE Int. Conf. Mechatron. Automat, pp. 1509–1514, Jun. 2006.
[14]W.C. Chang, C.K. Shao, “Hybrid eye-to-hand and eye-in-hand visual servoing for autonomous robotic manipulation” SICE Annual Conference 2010, Page(s): 415-422.
[15]A. Muis and K. Ohnishi, “Eye-to-hand approach on eye-in-hand con&;#64257;guration within real-time visual servoing,” in Proc. IEEE Int. Work-shop Adv. Motion Control, 2004, pp. 647–652.
[16]Dijkstra, E. W. (1959). "A note on two problems in connexion with graphs". Numerische Mathematik 1: 269–271.
[17]N. Otsu, “A threshold selection method from gray-level histograms”. IEEE Trans. Sys., Man., Cyber. 9: 62–66 (1979)
[18]Y. H. Liu, H. Wang, C. Wang and K. K. Lam, “Uncalibrated Visual Servoing of Robots Using a Depth-Independent Interaction Matrix,” IEEE Transactions on Robotics, Vol. 22, No. 4, pp. 804-817, 2006.
[19]H. Li, J. Liu, Y. Li. “Visual Servo of Uncalibrated Eye-in-hand System with Time-delay Compensation”. Proceedings of the 8th World Congress on Intelligent Control and Automation. 2010.
[20]F. Chaumette, S. Hutchinson, “Visual Servo Control Part I: Basic Approaches”, IEEE Robotics &; Automation Magazine, vol.14, issue 1, pp. 109-118, 2006.
[21]C. Cai, E. Dean-Leon, D. Mendoza, N. Somani, A. Knoll, “Uncalibrated 3D stereo image-based dynamic visual servoing for robot manipulators,” Intelligent Robots and Systems (IROS), 2013 IEEE/RSJ
[22]Y. Zhao, W.-F. Xie, and X.-W. Tu, “Multiple cameras visual servoing used for large scale 3d positioning,” in Intelligent Control and Automation (WCICA), 2011 9th World Congress on, june 2011, pp. 462–467.
[23]S. Hutchinson, G. D. Hager, and P. I. Corke, “A tutorial on visual servo control,” IEEE Transactions on Robotics and Automation, vol. 12, pp. 651–670, Oct. 1996.
[24]Kalman, R. E. (1960). "A New Approach to Linear Filtering and Prediction Problems". Journal of Basic Engineering 82 (1): 35–45.
[25]R. Ahuja, J. B. Orlin, S. Pallottino, and M. G. Scutella, “Dynamic shortest path minimizing travel times and costs,”Networks, vol. 41, no. 4, pp. 197–205, 2003.
[26]B Xiao, J Cao, Q Lu, "Dynamic Shortest Path Tree Update for Multiple Link State Decrements, " The Journal of Supercomputing, Volume 46, Issue 3, pp 237-256, 2008
[27]G. Ecorchard and P. Maurine, “Self-calibration of delta parallel kinematic mechanism robots with elastic deformation compensation,” International Conference on Intelligent Robots and Systems, 2005.(IROS 2005), pp. 1283–1288.