本論文針對人形機器人研究，設計並實現了多功能視覺與控制系統。首先，詳細說明了一系列的人形機器人其硬體規格及感測器模組架構。接著提出了以FPGA為基礎且具有圖樣辨識及字母辨識功能的影像系統。影像資訊由CMOS影像感測器擷取，將顏色與雜訊做處理及濾除後，利用邊緣特徵點來分析所擷取到的影像。經由圖樣辨識功能，我們可在複雜環境中同時辨識多個幾何圖樣及箭頭方向與其指向角度。在字母辨識功能中，先利用影像切割法找出影像中有效的區域，再利用編解碼的方式找出相對應的字母，透過此演算法可同時辨識出在影像中的多個字母。除此之外更針對FIRA與RoboCup此二大主要國際機器人競賽，提出了具有PK控制器功能以及自我定位功能的影像系統。PK控制器主要用於FIRA賽事中，功能主要有目標物的追蹤及進攻策略的編制，並且我們將此控制器完全實現於FPGA發展板上。而自我定位功能主要是用於RoboCup挑戰賽事中。利用機器人可視範圍內的場地資訊及相對適應函數的制定，設計出一階層式粒子濾波器來完成此自我定位功能。
同時我們也提出了一種新的人形機器人步態動作設計及相對應的步態學習方法，再利用拉格朗日多項式插值法及模糊控制器的搭配，讓人形機器人能以不間斷的步行動作去追隨不同型態的目標物。首先我們將行走步態重新定義並將其參數化。然後透過策略梯度增強式學習得到快速且穩定的步態動作後，接著利用拉格朗日多項式插值法與模糊控制器的結合，在不同的步態動作中，合成出特定行走方向與速度的動作，讓機器人可以連續的步伐來追隨目標物。
在雙足機器人研究中，除了機器人步行速度外，機器人步行穩定度也是一個重要的課題。對於人形機器人步行穩定性的分析，我們也提出了一個新的ZMP軌跡模型及實際ZMP量測方法，讓使用者能針對不同機器人來設計出適合的ZMP軌跡及量測出正確的ZMP位置。除此之外我們利用卡曼濾波器及模糊控制器的使用，設計出一動態平衡控制系統。透過感測器資訊融合的方式，利用ZMP誤差及機器人身體的傾斜角度，作為動態平衡控制系統的輸入並利用模糊控制器的輸出，適當地調整機器人各關節的角度。讓人形機器人可透過此動態平衡控制，成功地步行於不同傾斜程度的地面情況。最後，相關的實驗結果均可驗證所提出的影像系統及控制系統，可有效地在人形機器人上實現。

In this dissertation, the system structure and the sensor modules of a series of humanoid robots are first described. Then, a FPGA-based vision system that consists of pattern recognition and character recognition is proposed. The image information is captured by the CMOS sensor, and the colors and noise are filtered out first. Then, the edge detection approach is utilized to find the feature points so as to analyze the image. Via pattern recognition, several geometric figures and the direction of the arrows can be recognized in a complex environment at the same time. For the character recognition, the image segmentation is applied to find the valid region, and the encoding method is used to sample the word. After the matching algorithm, multi-characters can be recognized at once.
For the international robot competitions, FIRA and RoboCup, the penalty kick controller and the localization method are also implemented in this vision system. The PK controller is developed for the PK event in HuroSot of FIRA, which mainly covers the development of the target tracking and the offence strategy. The computations, including the image processing and the fuzzy logic controller design, are operated on an FPGA board. With the constructed vision system, a self-localization method is also realized. To save computation time and increase the accuracy of the localization, a hierarchy particle filter is developed. Using the construction of a grid map for the playing field, the contribution of line, and the fitness function, the hierarchy particle filter is built up and employed in the technical challenges in RoboCup.
Besides, we propose the implementation of fuzzy motion control based on reinforcement learning and Lagrange polynomial interpolation (LPI) for gait synthesis of biped robots. The procedure of a walking gait is redefined into three states, and the parameters of this designed walking gait are determined. Then, the machine learning approach applied to adjusting the walking parameters is policy gradient reinforcement learning (PGRL), which can execute real-time performance and directly modify the policy without calculating the dynamic function. Given a parameterized walking motion designed for biped robots, the PGRL algorithm automatically searches the set of possible parameters and finds the fastest possible walking motion. The results show that the robot not only has more stable walking but its walking speed also increases after learning. LPI, moreover, is employed to transform the existing motions to the motion which has a revised angle determined by the fuzzy motion controller. Then, the biped robot can continuously walk in any desired direction through this fuzzy motion control.
In addition to the walking speed, the walking ability is a fundamental research for biped robots. The Zero Moment Point (ZMP) criterion is one of the useful standards for measuring biped robot walking. In this dissertation, a new ZMP trajectory model and an actual ZMP measurement method are proposed to modulate the ZMP trajectory both in sagittal and lateral planes. Furthermore, a dynamic balance control (DBC), which includes the Kalman filter and the fuzzy motion controller (FMC), is also designed to keep the body balance and follow the desired ZMP reference. Using the sensor fusion technique, ZMP error and trunk inclination measured by the force sensor and accelerometer serve as the inputs for FMC, which are presented to correct each joint of the biped robot dynamically. When a biped robot walks under different ground conditions, the coordination of the designed ZMP trajectory and proposed DBC can achieve a successful biped walking. Finally, the experimental results successfully validate the feasibility and effectiveness of the proposed multi-functional vision and control systems.

[1] O. Ayhan and K. Erbatur, “Biped robot walk control via gravity compensation techniques,” in Proc. 30th Annual Conference of IEEE Industrial Electronics Society, IECON 2004, vol. 1, pp. 621-626.
[2] S. T. Acton and D. P. Mukherjee, “Area operators for edge detection,” Pattern Recognition Letters, vol. 21, no. 8, pp. 771-777, July 2000.
[3] B. Bödvarssona, S. Klima, M. Mørkebjerga, S. Mortensena, C. H. Yoona, J. Chena, J. R. Maclarena, P. K. Lutherb, J. M. Squirec, P. J. Bonesa, and R. P. Millane, “A morphological image processing method for locating myosin filaments in muscle electron micrographs,” Image and Vision Computing, vol. 26, pp. 1073-1080, Aug. 2008.
[4] J. P. Berrut and L. N. Trefethen, “Barycentric Lagrange Interpolation,” SIAM Review, vol. 46, pp. 501-517, 2004.
[5] A. Chemori, and A. Loria, “Control of a planar underactuated biped on a complete walking cycle,” IEEE Transactions on Automatic Control,, vol. 49, no. 5, pp. 838-843, May 2004.
[6] A. Cherubini, F. Giannone, L. Iocchi and P. F. Palamara, “An extended policy gradient algorithm for robot task learning,” in Proc. 2007 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS 2007, pp. 4121-4126.
[7] A. Cherubini, F. Giannone, L. Iocchi, M. Lombardo and G. Oriolo, “Policy gradient learning for a humanoid soccer robot,” Robotics and Autonomous Systems, vol. 57, pp. 808-818, Apr. 2009.
[8] C. T. Cheng, H. C. Chen, Y. Y. Hu, and C. C. Wong, “Fuzzy balancing control of a small-sized humanoid robot based on accelerometer,” International Journal of Fuzzy Systems, vol.11 no.3, pp. 146-153, Sep. 2009.
[9] K. C. Choil, H. J. Lee, and M. C. Lee, “Fuzzy posture control for biped walking robot based on force sensor for ZMP,” in Proc. SICE-ICASE 2006, International Joint Conference, 2006, pp. 1185-1189.
[10] R. G. Casey and E. Lecolinet, “A survey of methods and strategies in character segmentation,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 18, no. 7, pp. 690-706, July 1996.
[11] Y. Choi, D. Kim, Y. Oh, and B. J. You, “Posture/Walking control for humanoid robot based on kinematic resolution of CoM jacobian with embedded motion,” IEEE Transactions on Robotics, vol. 23, no. 6, pp. 1285-1293, Dec. 2007.
[12] Y. Choi, B. J. You, and S. R. Oh, “On the stability of indirect ZMP controller for biped robot systems,” in Proc. 2004 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS 2004, vol. 2, pp. 1966-1971.
[13] A. Dasgupta and Y. Nakamura, “Making feasible walking motion of humanoid robots from human motion capture data,” in Proc. IEEE International Conference on Robotics and Automation, 1999, pp. 1044-1049.
[14] A. Doucet, N. De Freitas, and N. Gordon, Sequential Monte Carlo Methods in Practice: Springer Verlag, 2001.
[15] K. Erbatur and O. Kurt, “Natural ZMP trajectories for biped robot reference generation,” IEEE Transactions on Industrial Electronics, vol. 56, no. 3, pp. 835-845, Mar. 2009.
[16] C. Fu and K. Chen, “Gait synthesis and sensory control of stair climbing for a humanoid robot,” IEEE Transactions on Industrial Electronics, vol. 55, no. 5, pp. 2111-2120, May 2008.
[17] J. S. Farrokh and M. Mohammed, “A Kalman-filter-based method for pose estimation in visual servoing,” IEEE Transactions on Robotics, vol. 26, no. 5, pp. 939-947, Oct. 2010.
[18] K. S. Fu, R. C. Gonzales, and C. S. G. Lee, Robotics: Control, Sensing, Vision, and Intelligence, McGraw-Hill Book Company, 1987.
[19] G. Grisetti, C. Stachniss, and W. Burgard, “Improved techniques for grid mapping with Rao-Blackwellized particle filters,” IEEE Transactions on Robotics, vol. 23, pp. 34-46, 2007.
[20] J. Y. Gil and R. Kimmel, “Efficient dilation, erosion, opening, and closing algorithms,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 24, Issue 12, pp. 1606-1617, Dec. 2002.
[21] T. Gevers, “Robust segmentation and tracking of colored objects in video,” IEEE Transactions on Circuits and Systems for Video Technology, vol. 14, pp. 776-781, June 2004.
[22] E. P. Hanavan, A Mathematical Model of the Human Body vol. 32: Aerospace Medical Research Laboratories, Aerospace Medical Division, Air Force Systems Command, 1964.
[23] H. P. Huang, J. L. Yan and T. H. Cheng, “Development and fuzzy control of a pipe inspection robot,” IEEE Transactions on Industrial Electronics, vol. 57, no. 3, pp. 1088-1095, Mar. 2010.
[24] I. Harmati and K. Skrzypczyk, “Robot team coordination for target tracking using fuzzy logic controller in game theoretic framework,” Robotics and Autonomous Systems, vol. 57, pp. 75-86, 2009.
[25] Z. Hou and T. S. Koh, “Robust edge detection,” Pattern Recognition, vol. 36, no. 9, pp. 2083-2091, Sep. 2003.
[26] S. Ito, Y. Aoyama, and H. Kawasaki, “A static balance control under periodic external force,” in Proc. SICE 2003 Annual Conference, 2003, vol. 2, pp. 1967-1972.
[27] J. G. Juang, “Fuzzy neural network approaches for robotic gait synthesis,” IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics, vol. 30, no. 4, pp. 594-601, Aug. 2000.
[28] A. L. Kun and T. W. Miller, “Control of variable-speed gaits for a biped robot,” IEEE Robotics & Automation Magazine, pp. 19-29, Sep. 1999.
[29] B. Kim, J. Park, S. Park and S. Kang, “Impedance learning for robotic contact tasks using natural actor-critic algorithm,” IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics, vol. 40, no. 2, pp. 433-443, Apr. 2010.
[30] J. Kwon and F. C. Park, “Natural movement generation using hidden Markov models and principal components,” IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics, vol. 38, no. 5, pp. 1184-1194, Oct. 2008.
[31] N. Kohl and P. Stone, “Policy gradient reinforcement learning for fast quadrupedal locomotion,” in Proc. IEEE International Conference on Robotics and Automation, 2004, vol. 3, pp. 2619-2624.
[32] S. Kajita, F. Kahehiro, K. Kaneko, K. Fujiwara, K. Harada, K. Yokoi, and H. Hirukawa, “Biped walking pattern generation using preview control of the zero-moment-point,” in Proc. IEEE International Conference on Robotics and Automation, 2003, vol. 2, pp. 1620-1626.
[33] B. J. Lee, D .Stonier, Y. D. Kim, J. K. Yoo and J. H. Kim, “Modifiable walking pattern of a humanoid robot by using allowable ZMP variation,” IEEE Transactions on Robotics, vol. 24, no. 4, pp. 917-925, Aug. 2008.
[34] C. K. Lin, “Fuzzy-basis-function-network-based tracking control for robotic manipulators using only position feedback,” IEEE Transactions on Fuzzy Systems, vol. 17, no. 5, pp. 1208-1216, Oct. 2009.
[35] C. M. Lin and C. H. Chen, “Robust fault-tolerant control for a biped robot using a recurrent cerebellar model articulation controller,” IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics, vol. 37, pp. 110-123, Feb. 2007.
[36] H. K. Lum, M. Zribi, and Y. C. Soh, “Planning and control of a biped robot,” International Journal of Engineering Science, vol. 37, no. 2, pp. 1319-1349, 1999.
[37] S. H. Liu, Design and Implementation of a Gait Pattern Generator Based on Genetic Algorithms and Fuzzy Control for Small-sized Humanoid Robot by Using SOPC, Master Thesis, National Cheng Kung University, July 2008.
[38] T.-H. S. Li, Y.-T. Su, S.-W. Lai, and J.-J. Hu, “Walking motion generation, Synthesis, and control for biped robot by using PGRL, LPI and fuzzy logic,” IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics, vol. 41, pp. 736-748, June, 2011.
[39] Z. Liu, C. Li, and W. Xu, “Hybrid control of biped robots in the double-support phase via H∞ approach and fuzzy neural networks,” IEE Proceedings Control Theory and Applications, 2003, vol. 150, no. 4, pp. 347-354.
[40] Z. Liu, Z. Yun, and Y. Wang, “A type-2 fuzzy switching control system for biped robots,” IEEE Transactions on Systems, Man, and Cybernetics, Part C: Applications and Reviews, vol. 37, no. 6, pp. 1202-1213, Nov. 2007.
[41] X. Mu and Q. Wu, “On impact dynamics and contact events for biped robots via impact effects,” IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics, vol. 36, no. 6, pp. 1364-1372, Dec. 2006.
[42] E. Ohashi, T. Sato, and K. Ohnishi, “A walking stabilization method based on environmental modes on each foot for biped robot,” IEEE Transactions on Industrial Electronics, vol. 56, Issue 10, pp. 3964-3974, Oct. 2009.
[43] J. H. Park and Y. K. Rhee, “ZMP trajectory generation for reduced trunk motions of biped robots,” in Proc. 1998 IEEE/RSJ International Conference on Intelligent Robots and Systems, , 1998, vol. 1, pp. 90-95.
[44] J. H. Park and E. S. Kim, “Foot and body control of biped robots to walk on irregularly protruded uneven surfaces,” IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics, vol. 39, no. 1, pp. 289-297, Feb. 2009.
[45] S. Park, Y. Han, and H. Hahn, “Fuzzy controller based biped robot balance control using 3D image,” International Journal of Fuzzy Systems, vol.11 no.3, pp. 202-212, Sep. 2009.
[46] T. Rofer, T. Laue, and D. Thomas, “Particle-filter-based self-localization using landmarks and directed lines,” in Proc. RoboCup 2005: Robot Soccer World Cup IX, 2006, pp. 608-615.
[47] A. W. Salatian, K. Y. Yi and Y. F. Zheng, “Reinforcement learning for a biped robot to climb sloping surfaces,” Journal of Robotic Systems, vol.14, no. 4, pp. 283-296, 1997.
[48] C. Su and R. M. Haralick, “Recursive erosion, dilation, opening, and closing transforms,” IEEE Transactions on Image Processing, vol. 4, Issue 3, pp. 335-345, Mar. 1995.
[49] Richard S. Sutton and Andrew G. Barto, Introduction to Reinforcement Learning, MIT Press, Cambridge, MA, 1998.
[50] K. Suwanratchatamanee, M. Matsumoto, and S. Hashimoto, “Haptic sensing foot system for humanoid robot and ground recognition with one leg balance,” IEEE Transactions on Industrial Electronics, Digital Object Identifier: 10.1109/TIE.2009.2030217.
[51] P. Sardain and G. Bessonnet, “Zero moment point-measurements from a human walker wearing robot feet as shoes,” IEEE Transactions on Systems, Man and Cybernetics, Part A: Systems and Humans, vol. 34, pp. 638-648, Sep. 2004.
[52] R. S. Sutton, D. McAllester, S. Singh, Y. Mansour, “Policy gradient methods for reinforcement learning with function approximation,” Advances in neural information processing systems, vol. 12, 2000.
[53] S. Suthaharan, “Image and edge detail detection algorithm for object-based coding,” Pattern Recognition Letters, vol. 21, pp. 549-557, June 2000.
[54] T. Sugihara, Y. Nakamura, and H. Inoue, “Real-time humanoid motion generation through ZMP manipulation based on inverted pendulum control,” in Proc. IEEE International Conference on Robotics and Automation, 2002, vol. 2, pp. 1404-1409.
[55] A. Takanish, M. Tochizawa, H. Karaki, and I. Kato, “Dynamic biped walking stabilized with optimal trunk and waist motion,” in Proc. IEEE/RSJ International Workshop on Intelligent Robots and Systems, 1989, pp. 187-192.
[56] T. Takagi and M. Sugeno, “Fuzzy identification of systems and its application to modeling and control,” IEEE Transactions on Systems, Man and Cybernetics, vol. 15, no. 1, pp. 116-132, 1985.
[57] B. Uorovac, M. Vukobratovic, and D. Suda, “An approach to biped control synthesis,” Robotica, vol. 7, pp. 231-241, 1989.
[58] P. R. Vundavilli and D. K. Pratihar, “Soft computing-based gait planners for a dynamically balanced biped robot negotiating sloping surfaces,” Applied Soft Computing, vol. 9, no. 1, pp. 191-208, Jan. 2009.
[59] G. Welch and G. Bishop, An Introduction to the Kalman Filter, University of North Carolina, Department of Computer Science, TR 95-041.
[60] J. Wallen, S. Gunnarsson, R. Henriksson, S. Moberg, and M. Norrlof, “ILC applied to a flexible two-link robot model using sensor-fusion-based estimates,” in Proc. 48th IEEE Conference on Decision and Control, 2009, pp.458-463.
[61] S. H. Won, F. Golnaraghi, and W. W. Melek, “A fastening tool tracking system using an IMU and a position sensor with Kalman filters and a fuzzy expert system,” IEEE Transactions on Industrial Electronics, vol. 56, no. 5, pp. 1782-1792, May 2009.
[62] S. H. Won, W. W. Melek, and F. Golnaraghi, “A Kalman/Particle filter-based position and orientation estimation method using a position sensor/inertial measurement unit hybrid system,” IEEE Transactions on Industrial Electronics, vol. 57, no. 5, pp. 1787-1798, May 2010.
[63] J. X. Xu and W. Wang, “A general internal model approach for motion learning,” IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics, vol. 38, no. 2, pp. 477-487, Apr. 2008.
[64] Y. H. Yu and C. C. Chang, “A new edge detection approach based on image context analysis,” Image and Vision Computing, vol. 24, no. 10, pp. 1090-1102, Oct. 2006.
[65] C. Zhu, Y. Tomizawa, X. Luo, and A. Kawamura, “Biped walking with variable ZMP, frictional constraint, and inverted pendulum model,” in Proc. IEEE International Conference on Robotics and Biomimetics, 2004, pp. 425-430.
[66] L. A. Zadeh, “Fuzzy sets,” Information and control, vol. 8, pp. 338-353, 1965.
[67] AGB65_4FS, [Online]. Available: http://www.robotsfx.com/robot/AGB65_4FS.html
[68] Altera, [Online]. Available: http://www.altera.com/
[69] Arduino, [Online]. Available: http://www.arduino.cc/
[70] CMOS-EYE, [Online]. Available: http://www.robotsfx.com/robot/CMOS-EYE.html
[71] FIRA, http://www.fira.net/
[72] FlyCAM-SD, [Online]. Available: http://www.lifeview.com.tw/
[73] Hitachi H48C, [Online]. Available: http://www.parallax.com/tabid/768/ProductID/97/Default.aspx
[74] IDG300, [Online]. Available: http://invensense.com/mems/gyro/idg300.html
[75] KONDO, [Online]. Available: http://kondo-robot.com/sys/
[76] MAXON, [Online]. Available: http://www.maxonmotor.com/
[77] Memsic 2125, [Online]. Available: http://www.parallax.com/tabid/768/ProductID/93/Default.aspx
[78] Pico820, [Online]. Available: http://axiomtek.com/
[79] QuickCam Pro5000, [Online]. Available: http://www.logitech.com/en-us/435/243
[80] RoboCup, http://www.robocup.org/
[81] ROBOTIS, [Online]. Available: http://www.robotis.com/
[82] Webcam C905, [Online]. Available: http://www.logitech.com/en-us/webcam-communications/
[83] ZIG-100, [Online]. Available: http://www.nodna.com/download/ROBOTIS/ZIG-100(english).pdf/