本篇研究提出以模仿的方式教導機器人學習事物的方法，使人們能以簡單的方式傳達所想表達的行為。相較於一般由機器人專家為各個機器人分別設定的方式，此種方法更適用於服務型機器人上，以教導人們日常生活中的各項工作。而本篇研究著重於如何讓機器人透過觀察的方式學習人類的行為，並且導入生物學習的概念，探討當生物面臨一項新事件時，所可能採用的學習模式。
本篇研究採用Robotis公司所發展的Bioloid機器人作為展示的平台，探討當機器人觀察表演者行為之後，如何將表演者完整的展示動作，並且能以過去學習資訊做為輔助，將工作做有效的分解，使其無須學習多餘的工作。將每項新行為根據其複雜程度，分別提出簡易型行為學習方法以及複雜型行為學習方法。在學習的方法上本篇研究，將傳統的運動學問題進行編碼，使其能在一般演化式計算上運作，並導入過去學習資訊做為輔助與變動型區域搜尋的方法，探討一般解決複雜問題所使用分割征服學習的差異性。
在一般的模仿學習裡，主要步驟分為如何辨識行為以及如何產生動作。本篇研究裡採用行為辨識方法，將冗長的工作做有效的分解，使各項子工作可採用簡易型行為學習方法。若其行為複雜程度過高時，則可採用本篇研究所提出的複雜行為學習方法，或可採用一般分割征服法使問題複雜度降低。因此，透過以上方法使得模仿學習，能夠以逐步簡化問題的方法做有效的學習。

This thesis presents an imitation-based methodology, also a simple and easy way, for a service robot to learn the behaviors demonstrated by the user. With this proposed method, a robot can learn human behavior through observation. Inspired by the concept of biological learning, this learning model is initiated when facing a new learning event. A series of experiments are conducted to use a humanoid robot as a platform to implement the proposed algorithm. Discussions are made of how the robot generates a complete behavior sequences performed by its demonstrator. Because it is time consuming for a robot to go through the whole process of learning, we thus propose a decomposed learning method to enhance the learning performance, that is, based on the past learning information, the robot can skip learning again the behaviors already known.
For simple robot behaviors, a hierarchical evolutionary mechanism is developed to evolve the complete behavior trajectories. For complex behaviors sequences, different ways are used to tackle the scalability problem, including decomposing the overall task into several sub-tasks, exploiting behavior information recorded previously, and constructing a new strategy to maintain population diversity. To verify our approach, a different series of experiments have been conducted. The results show that our imitation-based approach is a natural way to teach the robot new behaviors. This evolutionary mechanism successfully enables a humanoid robot to perform the behavior sequences it learns.

圖目錄 VII
表目錄 IX
第一章 緒論 1
1.1 研究背景 1
1.2 研究動機與目的 2
1.3 論文架構 3
第二章 文獻探討 4
2.1 機器人學習(Robot Learning) 4
2.2 基因演算法(Genetic Algorithm) 5
2.2.1 交配與突變操作 7
2.2.2 基因演算應用於機器人 9
2.3 模仿學習(Imitation Learning) 11
2.3.1 數學模型學習方法 11
2.3.2 統計模型學習方法 12
2.3.3 模組基底學習方法 13
2.4 姿勢辨識 14
2.4.1 隱藏式馬可夫鏈(Hidden Markov Chain) 15
2.4.2 粒子濾波器(Particle Filtering) 16
2.4.3 有限狀態機(Finite State Machine) 16
2.4.4 軟式計算(Soft Computing) 17
第三章 機器人系統架構與硬體規格 18
3.1 人機互動介面 18
3.2 硬體規格 19
3.2.1 馬達(Dynamixal AX-12) 19
3.2.2 無線通訊模組(Zig-100) 20
第四章 機器人模仿學習研究方法 22
4.1 工作定義 23
4.2 工作分割 24
4.2.1 工作分割演算法 25
4.3 工作生成 29
4.3.1 染色體表示方式 30
4.3.2 評估函數 30
4.3.3 簡易行為學習 32
4.3.4 複雜行為學習 32
4.3.5 變動型區域學習演算法 33
第五章 實驗結果 38
5.1 行為辨識 38
5.2 基本組件 38
5.2.1 擷取動作 40
5.3 動作產生 44
5.3.1 簡易行為學習 44
5.3.2 複雜行為學習 53
5.3.2.1不同族群大小情況 56
5.3.2.2 加入資訊學習效果 58
5.3.2.3 變動型區域學習效果 61
5.3.2.4 變動型區域學習加入資訊學習效果 65
5.3.2.5 工作拆解分割學習效果 71
第六章 結論與未來研究 75
6.1 研究結論 75
6.2 未來展望 76
參考文獻 77

[1] Aladjov, H. T., & Raikova, R. T. (2002). Hierarchical Genetic Algorithm Versus Static Optimization-investigation of Elbow Flexion and Extension Movements. Journal of Biomechanics, vol. 35 , pp. 1123-1135.
[2] Arulampalam, S. M., Maskell, S., Gordon, N., & Clapp, T. (2002). A Tutorial on Particle Filters for Online Nonlinear/Non-Gaussian Bayesian Tracking. IEEE Transactions on　Signal Processing, vol. 50 , pp. 174-188.
[3] Billard, A., & Calinon, S. (2007). Incremental Learning of Gestures by Imitation in a Humanoid Robot. Proceedings of the ACM/IEEE International Conference on Human-Robot Interaction , pp. 255-262.
[4] Calinon, S., Guenter, F., & Billard, A. (2007). On Learning, Representing and Generalizing a Task in a Humanoid Robot. IEEE Transactions on Systems, Man, and Cybernetics,Part B, vol. 37 , pp. 286-298.
[5] Carlos, A. A., Rajesh, M. E., Hu, L., Zhou, C., & Hu, H. (2009). Generating Human-like Soccer Primitives from Human Data. Robotics and Autonomous System, vol. 57 , pp. 860-869.
[6] Chalup, S. K., Murch, C. L., & Quinlan, M. J. (2007). Machine Learning With AIBO Robots in the Four-Legged League of RoboCup. IEEE Transactions on Systems, Man, and Cybernetics, Part C: Applications and Reviews, vol. 27 , pp. 297-310.
[7] Donald, M. (1997). Origins of the Modern Mind: Three Stages in the Development of Culture and Congnition. London: Springer-Verlag.
[8] Frenkel, M., & Basr, R. (2003). Curve Matching Using the Fast Marching Method. EMMCVPR , pp. 35-51.
[9] Holland, J. (1975). Adaptation in Natural and Artificial Systems. Cambridge, MA: MIT.
[10] Juan, M. A., Bessiere, P., & Mazer, E. (1993). Using Genetic Algorithms for Robot Motion Planning. Geometric Reasoning for Perception and Action, vol. 708 , pp. 84-93.
[11] KÄÄRIÄINEN, M. (2006). Active Learning in the Non-realizable Case. Proceedings of the 17th International Conference on Algorithmic Learning Theory, vol. 4264 , pp. 63-77.
[12] Kitano, M. F. (1998). Development of an Autonomous Quadruped Robot for Robot Entertainment. Autonomous Robots, vol. 5 , pp. 7-18.
[13] Mataric, M. (2000). Getting Humanoids to Move and Imitate. IEEE Intelligent Systems and Their Application, vol. 15 , pp. 18-24.
[14] Mataric, M., & Cliff, D. (1996). Challenges in Evolving Controllers for Physical Robot. Robotics and Autonomous Systems, vol. 19 , pp. 67-83.
[15] Min, A., Taura, T., & Shiose, T. (2007). A Study on Acquiring Underlying Behavioral Criteria for Manipulator Motion by Focusing on Learning Efficiency. IEEE Transactions on Systems, Man and Cybernetics, Part A, vol. 37 , pp. 445-455.
[16] Mitra, S., & Acharya, T. (2007). Gesture Recognition: A survey. IEEE Transactions on Systems, Man, and Cybernetics, Part C: Applications and Reviews, vol. 37 , pp. 311-324.
[17] NakaoKa, S., Nakazawa, A., Kanehiro, F., Kaneko, K., Morisawa, M., Hirukawa, H., et al. (2007). Learning from Observation Paradigm: Leg Task Models for Enabling a Biped Humanoid Robot to Imitate Human Dances. The International Journal of Robotics Research, vol. 26 , pp. 829-844.
[18] Nehaniv, C., & Dautenhahn, K. (2000). Of Hummingbirds and Helicopters: An Algebraic Framework for Interdisciplinary Studies of Imitation and Its Applications. Interdisciplinary Approaches to Robot Learning, vol. 24 , pp. 136-161.
[19] Rohlfing, K. J., & Yukie, N. (2009). Computational Analysis of Motionese Toward Scaffolding Robot Action Learning. IEEE Transactions on Autonomous Mental Development, vol. 1 , pp. 44-54.
[20] Saunders, J., Dautenhahn, K., & Chrystopher, N. L. (2006). Teaching Robots by Moulding Behavior and Scaffolding the Environment. Proceedings of the 1st ACM SIGCHI/SIGART Conference on Human-Robot Interaction , pp. 118-125.
[21] Sebag, M., & Schoenauer, M. (1997). A Society of Hill-climbers. Proceedings of IEEE International Conference on Evolutionary Computation, 1997 , pp. 319-324.
[22] Wilson, A. D., & Bobick, A. F. (1999). Parametric Hidden Markov Models for Gesture Recognition. IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 21 , pp. 884-900.