請用此 Handle URI 來引用此文件：
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/57124完整後設資料紀錄
| DC 欄位 | 值 | 語言 |
|---|---|---|
| dc.contributor.advisor | 羅仁權(Ren C. Luo) | |
| dc.contributor.author | Cheng-Hsun Hsieh | en |
| dc.contributor.author | 謝政勳 | zh_TW |
| dc.date.accessioned | 2021-06-16T06:35:32Z | - |
| dc.date.available | 2014-08-12 | |
| dc.date.copyright | 2014-08-12 | |
| dc.date.issued | 2014 | |
| dc.date.submitted | 2014-08-01 | |
| dc.identifier.citation | [1] D. Stewart, “A platform with six degrees of freedom”, Proceedings of the IMechE, 1965.
[2] P. Maurine, E. Dombre, “A calibration procedure for the parallel robot Delta 4”, International Conference on Robotics and Automation, 1996. [3] J.M. Perez, S. Schreiner, G.E. Gorton, “Evaluation of the VITUS Smart Laser Scanner for Accuracy, Resolution and Repeatability for Clinical Assessment of Pectus Deformities and Scoliosis”, Bioengineering Conference on Proceedings of the IEEE 32nd Annual Northeast, 2006. [4] J. Denavit, R. S. Hartenberg. “A kinematic notation for lower-pair mechanisms based on matrices”, Trans. ASME, 1965. [5] T.W. Wright, “Elements of Mechanics Including Kinematics, Kinetics and Statics”, London, E. and F.N. Spon, 1896. [6] R.P. Paul, “Robot manipulators: mathematics, programming, and control: the computer control of robot manipulators”, The MIT Press, 1981. [7] S. James, “Calculus: Early Transcendentals (6th ed.)”, Cengage Learning, 6 edition, 2008. [8] I.V. Tetko, D.J. Livingstone, A.I. Luik, “Neural network studies. 1. Comparison of Overfitting and Overtraining”, J. Chem. Inf. Comput., 1995. [9] T. Niaritsiry, N. Fazenda, R. Clavel, “Study of the sources of inaccuracy of a 3 DOF Flexure Hinge based Parallel Manipulator”, Proceedings of the IEEE International Conference on Robotics and Automation (ICRA), 2004. [10] L.W. Tsai, “Robot Analysis: The Mechanics of Serial and Parallel Manipulators”, 1999. [11] F.X. Rosenblatt, “Principles of Neurodynamics: Perceptrons and the Theory of Brain Mechanisms”, Washington DC, 1961. [12] D.E. Rumelhart, G.E. Hinton, R.J. Williams, “Learning Internal Representations by Error Propagation”, The MIT Press, 1986. [13] X. Yao, ” Evolving artificial neural networks”, Proceedings of the IEEE, 1999. [14] M. Mohri, A. Rostamizadeh, A. Talwalkar, “Foundations of Machine Learning,” The MIT Press, 2012. [15] Universal Robots Inc. http://www.fanuc.co.jp/eindex.htm. [16] Universal electronics components Inc. http://www.deltaww.com/default.aspx?hl=zh-TW. [17] Delta robot form FAUNC. http://www.robots.com/fanuc/m-3ia. [18] Programmable Logic Controller. http://en.wikipedia.org/wiki/Programmable_logic_controller. [19] T.S. Yan, “An Improved Genetic Algorithm and Its Blending Application with Neural Network”, 2nd International Workshop on Intelligent Systems and Applications (ISA), 2010. [20] T. Sanguanchue, K. Jearanaitanakij, “Hybrid algorithm for training feed-forward neural networks using PSO-information gain with back propagation algorithm”, 9th International Conference on Electrical Engineering/Electronics, Computer, Telecommunications and Information Technology (ECTI-CON), 2012. [21] J.J. Rodriguez, J.K. Aggarwal, “Quantization error in stereo imaging”, Computer Society Conference on Computer Vision and Pattern Recognition(CVPR), 1988. [22] D. Zhang, Z. Gao, B. Song, Y. J. Ge, “Configuration design and performance analysis of a multidimensional acceleration sensor based on 3RRPRR decoupling parallel mechanism,” IEEE Conference on Chinese Control Conference (CCC), 2009. [23] S. Lessard, P. Bigras, and I. A. Bonev, “A new medical parallel robot and its static balancing optimization,” ASME Journal of Medical Devices, 2007. [24] J. P. Merlet, “Optimal design for the micro parallel robot MIPS,” IEEE International Conference on Robotics and Automation (ICRA), 2002. [25] T.Y. Gao, Meng. Li, “Calibration method and experiment of Stewart platform using a laser tracker,” IEEE International Conference on Systems, Man and Cybernetics (ICSMC), 2003. [26] Y. Meng, H. Zhuang, “Autonomous robot calibration using vision technology,” Robotics and Computer-Integrated Manufacturing, 2007. [27] N. Andreff, P. Martinet, “Vision-based self-calibration and control of parallel kinematic mechanisms without proprioceptive sensing,” Intelligent Service Robotics, 2009. [28] X. Ren, Z. Feng, C. Su, “A new calibration method for parallel kinematics machine tools using orientation constraint,” International Journal of Machine Tools and Manufacture, 2009. [29] O. Masory, “Kinematic calibration of a Stewart platform using pose measurements obtained by a single theodolite,” IEEE International Conference on Intelligent Robots and Systems (IROS), 1995. [30] G. Canepa, J.M. Hollerbach, A. Boelen, “Kinematic calibration by means of a triaxial accelerometer,” IEEE International Conference on Robotics and Automation (ICRA), 1994. [31] A. Traslosheros, J.M. Sebastián, E. Castillo, F. Roberti, R. Carelli, “One camera in hand for kinematic calibration of a parallel robot”, IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2010. [32] P. Maurine, E. Dombre, “A calibration procedure for the parallel robot Delta 4,” IEEE International Conference on Robotics and Automation (ICRA), 1996. [33] P. Renaud, N. Andreff, F. Marquet, P. Martinet, “Vision-based kinematic calibration of a H4 parallel mechanism,” IEEE International Conference on Robotics and Automation (ICRA), 2003. [34] D. Zhang, Z. Gao, “Optimal Kinematic Calibration of Parallel Manipulators With Pseudoerror Theory and Cooperative Coevolutionary Network,” IEEE Transactions on Industrial Electronics (TIE), 2012. [35] C.S. Yee, K.B. Lim, “Forward kinematics solution of Stewart platform using neural networks,” Neurocomputing, 1997. [36] S. Liu, W.L. Li, Y.C. Du, F. Liang “Forward kinematics of the Stewart platforms using hybrid immune genetic algorithm,” IEEE International Conference on Mechatronics and Automation (ICMA), 2006. [37] D. Wang, Y. Bai, “Calibration of Stewart platforms using Neural Networks,” IEEE Conference on Evolving and Adaptive Intelligent Systems (EAIS), 2012. [38] N. Fazenda, E. Lubrano, S. Rossopoulos, and R. Clavel, 'Calibration of the 6 DOF high-precision flexure parallel robot “Sigma 6”,' Proceedings of 5th Parallel Kinematics Seminar. Chemnitz, Germany, 2006. [39] J. Sola, J. Sevilla, “Importance of input data normalization for the application of neural networks to complex industrial problems”, IEEE Transactions on Nuclear Science, 1997. [40] A. Malinowski, J.M. Zurada, P.B. Aronhime, “Minimal training set size estimation for neural network-based function approximation”, IEEE International Symposium on Circuits and Systems(ISCAS), 1994. [41] J.B. Anthony, V.W. Mladen, H.S. Harold, “Training data requirement for a neural network to predict aerodynamic coefficients”, Independent Component Analyses, Wavelets, and Neural Networks, 2003. [42] M. Turmon, F. Terrence, “Sample Size Requirements for Feedforward Neural Networks”, Advances in Neural Information Processing Systems, 1995. [43] S. Hyontai, “The effect of training set size for the performance of neural networks of classification”, World Scientific and Engineering Academy and Society (WSEAS) on computers, 2010. [44] The Delta robot. http://ppt.cc/Pvrz. [45] The repeatability, accuracy, and resolution. http://ppt.cc/VfOq. [46] 彭彥嘉, “並聯式三軸機器人位置運動學分析”, 機械工業雜誌, 329期2010年8月號, p084. | |
| dc.identifier.uri | http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/57124 | - |
| dc.description.abstract | 校準技術已被廣泛應用於機器人，校正為機器工具的準確度提供了具有成本效益的解決方案。標準的校正過程中，需進行三種不同的工作程序：建模，量測和校準。一般採用運動學模型建模，而對並聯式機器人來說是建立反向運動學模型。量測的方式一般採用高精度儀器對末端點做位置量測，收集一組完整的三維數據，建立系統的誤差模型。量測後就是校準，一般使用兩種不同的技術進行校準。
第一種方法稱為“基於模型的方法”，要求在電動機與機械手的操作坐標系之間建立一個校準關係。這個關係需從機器人的幾何形狀、環境的誤差來源考慮近似的數學函數與係數參數。高精度多自由度的並聯機器人一般採用多項式模擬誤差模型，再對此誤差模型進行誤差補償與校準。 第二種方法稱為“無模型的方法”，此方法中使用者並不需要知道影響機器人精度的主要誤差源，而是採用學習的方式去補償誤差。無模型的方法一般採用人工神經網絡來實現。以往神經網路的使用方式需要使用者反覆的調適網路架構，才有辦法得到好的學習結果。本論文提出讓使用者不需事先給定神經網路的架構，只需給定神經網路輸入與輸出，即可根據訓練的資料得出優化過後的網路架構，提高訓練效率並增加精度。與其他優化的神經網路比較，本論文提出的基於決策樹和協同網路的導傳遞算法得到更好的結果，不僅提高了方便性，也增加了準確性。此外我們提出的雙眼視覺系統輔以校正版，大大的減低了以往的雷射校正系統昂貴的負擔。 這種技術對工業應用帶來廣泛的益處，利益在於該機器人在實際的微加工操作上透過軟體的技術去修正誤差，減少從硬體端修正誤差所帶來的龐大成本。 | zh_TW |
| dc.description.abstract | Calibration technique has been widely used in robotics, and it provides high accuracy and cost-effective solution for the robot. The calibration procedure only modifies the programming part instead of the hardware of robot design or tightening the manufacturing tolerances. This thesis proposed the highly accurate calibration method for the multi-DoF parallel robot.
There are three steps in the standard calibration procedure: modeling, measurement, and correction. Kinematic modeling is a common way to modeling. Inverse kinematic model is used in parallel mechanism modeling. The high accuracy instrument is used to measure the position of end-effector. As for the position of end-effector, we can collect a complete set of 3D data and construct the error model. Correction starts after the measurement. The two different techniques are used in correction. The first method is model-based model. It constructs a correction relationship between the motor and operational coordinates of the robot. The correction relationship considers the geometric of robot and the environment error to decide the coefficients of mathematics function. The high accuracy multi-DoF parallel robot use polynomial function to construct the error model, and use this error model to compensate the error and calibration. The second method is model-free model. In this method, user doesn’t need to know the error source affecting the robot accuracy. The method compensates the error by learning method, and artificial neural network is commonly chosen for learning method. This thesis proposed an effective Visual Calibration System for Parallel Robot Using Cooperative Coevolutionary Network and Decision Tree Approach. The method can self-construct and optimize the neural network structure and parameters for the individual training set, and keeps the good prediction ability. This method combines with inverse kinematic model and it can find the accurate relationship between the motor and operational coordinates of the robot, and doesn’t need to consider the coefficient of polynomial and error model of the robot. The self-construction and optimization of the structure and parameters of neural network can help the robot achieve high accuracy in the workspace, and adapt in any source of error in the environment. The calibration technique brings a lot of benefit for the industrial applications, decrease the huge cost of error modification from the hardware and solves the problem using software technique. | en |
| dc.description.provenance | Made available in DSpace on 2021-06-16T06:35:32Z (GMT). No. of bitstreams: 1 ntu-103-R01921088-1.pdf: 4093107 bytes, checksum: 8d2d6f0a45a323f95c0a82223e607661 (MD5) Previous issue date: 2014 | en |
| dc.description.tableofcontents | Table of Contents
誌謝 i 中文摘要 ii Abstract iii Table of Contents v List of Figures vii List of Tables x Chapter 1 Introduction 1 1.1 Definitions 1 1.1.1 Mechanism 1 1.1.2 Accuracy, Resolution and Repeatability 3 1.1.3 Modeling Process 5 1.1.4 Sources of Error 6 1.1.5 Calibration Procedure 7 Chapter 2 LITERATURE SURVEY 9 2.1 Kinematic Modeling 9 2.1.1 Delta robot inverse kinematic model 9 2.1.2 Delta robot forward kinematics 14 2.2 Artificial neural network 18 2.2.1 Model of ANN 18 2.2.2 The Learning of Neural Network 23 2.2.3 Multiple layer perceptron 24 2.2.4 BP algorithm 25 2.2.5 The learning step of BP algorithm 32 Chapter 3 Hardware Structure 38 3.1 System Structure 38 3.1.1 Servo motor 39 3.1.2 Actuator 39 3.1.3 Reduction gears 40 3.1.4 Controller 41 3.1.5 Structure design 42 3.1.6 Main structure 43 3.1.7 The suck nozzle 43 3.1.8 The maximum load 44 3.1.9 Maximum Speed 45 3.1.10 The maximum working space 46 3.1.11 Electromechanical control 46 3.1.12 Cost 48 Chapter 4 Calibration of Delta Robot 49 4.1 Introduction 49 4.2 System Description 53 4.2.1 Possible error in the system 53 4.2.2 System structure 55 4.3 METHODOLOGIES 56 4.3.1 Manipulate the delta robot to every grids 56 4.3.2 Traditional kinematic calibration 59 4.3.3 DCCN calibration 60 Chapter 5 Experiment 80 Chapter 6 Conclusion and Contributions 88 Chapter 7 Future work 90 References 91 VITA 94 List of Figures Figure 1-1 The Stewart platform 2 Figure 1-2 The Delta robot 3 Figure 1-3 Sources of error commonly considered in static robot calibration problems 4 Figure 1-4 Sources of error commonly considered in static robot calibration problems 7 Figure 2-1 The structure of Delta robot 9 Figure 2-2 The inverse kinematic parameter of Delta robot 10 Figure 2-3 The inverse kinematic algorithm analysis of Delta robot 11 Figure 2-4 The forward kinematic parameter of Delta robot 15 Figure 2-5 The moving direction of 15 Figure 2-6 The three spheres cross the same point 16 Figure 2-7 unit of artificial neural network 19 Figure 2-8 Threshold function 20 Figure 2-9 Piecewise-Linear Function 21 Figure 2-10 Sigmoid function 22 Figure 2-11 Tangent hyperbolic function 22 Figure 2-12 Multiple layer perceptron 24 Figure 2-13 The working signal and the error signal in MLP 26 Figure 2-14 The two hidden layers BP network 27 Figure 3-2 The motor of the robot 39 Figure 3-3 The design of actuator 40 Figure 3-4 The design of reduction gear 41 Figure 3-5 UTC 400V/P four joint controller 41 Figure 3-6 The structure of Delta robot 42 Figure 3-7 The two rods design of the forearm. 43 Figure 3-8 The maximum load calculation 45 Figure 3-9 The maximum speed calculation 46 Figure 3-10 The maximum working space 46 Figure 3-11 The system of Delta robot 48 Figure 4-1 The different kinds of error 54 Figure 4-2 System Structure 55 Figure 4-3 Standard point and Nominal pose in one grid 57 Figure 4-4 The corner points and the position point 57 Figure 4-5 The moving direction of measurement 59 Figure 4-6 system structure of kinematic calibration 60 Figure 4-7 BP network structure 61 Figure 4-8 The combination of a neuron 62 Figure 4-9 The division of the sub-network 64 Figure 4-10 The coding of a two hidden layers MLP 66 Figure 4-11 The relative of binary coding and real value coding 67 Figure 4-12 The threshold coding 67 Figure 4-13 The crossover process 69 Figure 4-14 The special structure in crossover 71 Figure 4-15 The special case in crossover 71 Figure 4-16 Iteration 1 75 Figure 4-17 Iteration 2 76 Figure 4-18 Iteration 3 76 Figure 4-19 Iteration 4 77 Figure 4-20 Iteration 5 78 Figure 4-21 Flow chart of DCCN 79 Figure 5-1 The system mechanism (up left), Delta robot (up right), the two orthogonal checkerboards (down left), and the dual camera system (down right) 81 Figure 5-2 The standard point (left, yellow point) and the measured point (right, blue point) 83 Figure 5-3 The schematic error distribution 84 List of Tables Table 1 The delta robot specification 39 Table 2 The specification of actuator 40 Table 3 The specification of reduction gear 40 Table 4 The cost of Delta robot 48 Table 5 The polynomial for kinematic calibration 60 Table 6 The comparison of performance 86 Table 7 The comparison of standard deviation 86 Table 8 The comparison of improved percentage of error 86 Table 9 The comparison of average error 87 | |
| dc.language.iso | en | |
| dc.subject | 機器人校正 | zh_TW |
| dc.subject | 並聯式機器人 | zh_TW |
| dc.subject | 決策樹 | zh_TW |
| dc.subject | 類神經網路 | zh_TW |
| dc.subject | 協同進化網路 | zh_TW |
| dc.subject | robot calibration | en |
| dc.subject | sub-mm accuracy | en |
| dc.subject | parallel robots | en |
| dc.subject | kinematic mechanism | en |
| dc.subject | neural networks | en |
| dc.subject | decision tree | en |
| dc.subject | Cooperative Coevolutionary Network | en |
| dc.title | 基於協同進化網路與決策樹之高效視覺校正系統應用於並聯式機器人 | zh_TW |
| dc.title | Effective Visual Calibration System for Parallel Kinematic Mechanism Robot Using Decision Tree with Cooperative Coevolutionary Network Approach | en |
| dc.type | Thesis | |
| dc.date.schoolyear | 102-2 | |
| dc.description.degree | 碩士 | |
| dc.contributor.coadvisor | 陳俊宏(Chun-Hung Chen) | |
| dc.contributor.oralexamcommittee | 郭重顯(Chung-Hsien Kuo) | |
| dc.subject.keyword | 機器人校正,並聯式機器人,類神經網路,決策樹,協同進化網路, | zh_TW |
| dc.subject.keyword | robot calibration,sub-mm accuracy,parallel robots,kinematic mechanism,neural networks,decision tree,Cooperative Coevolutionary Network, | en |
| dc.relation.page | 94 | |
| dc.rights.note | 有償授權 | |
| dc.date.accepted | 2014-08-04 | |
| dc.contributor.author-college | 電機資訊學院 | zh_TW |
| dc.contributor.author-dept | 電機工程學研究所 | zh_TW |
| 顯示於系所單位： | 電機工程學系 | |
文件中的檔案：
| 檔案 | 大小 | 格式 | |
|---|---|---|---|
| ntu-103-1.pdf 未授權公開取用 | 4 MB | Adobe PDF |
系統中的文件，除了特別指名其著作權條款之外，均受到著作權保護，並且保留所有的權利。