臺灣博碩士論文加值系統

由於人工智慧立基人類之智慧行為，因此以代理人觀念及所處情境特性作為分析架構(如PEAS架構)是很自然的，引用及融入軟體代理人或機器人來貫通教學也很合理，不過將軟硬體結合並與情境中之真人互相合作則是新進的想法。2007年Wilensky提出真人、嵌入式系統與虛擬代理人中介平台(HEV-M)並發表實驗成果；不過，HEV-M架構雖達到人機互動，也應用於代理人教學，但平台架構本身及其應用系統缺乏分析及設計的軟體工程流程，效益驗證方面也缺乏團隊合作性之理論參考。
本研究在融入機器人於人工智慧教學的基礎上，於人機互動合作行為實驗的效益驗證與詮釋，參考Woods之共同任務與團隊玩家二大理論。於教學實驗系統之分析設計方法，則使用代理人導向軟體工程之Gaia方法論，並以MSRDS-VPL執行平台與 Lego NXT硬體機器人結合進行實作。我們進而應用此方法於人工智慧代理人單元教學課程，完成二階段遊戲實驗及驗證，第一階段複製Wilensky之實驗，並確認本方法之理論框架，系統設計方法與學生參與行為之分析與解釋之可行性，第二階段則確認本方法可擴充延伸Wilensky實驗，進一步分解人機團隊的綁固性，擴大參與模擬實驗模式為多機器人合作對抗實驗，發展更有趣的人機互動教案遊戲實驗。

The notion of agency and the characteristics of environments where agents situate as a conceptual and analytical framework are very natural, for Artificial Intelligence textbooks. As a result, its incorporation into the software agent or robot to run through the teaching is also very reasonable. However, the combination of hardware and software and the interaction with human beings in situations of co-operation becomes a focus recently. Wilensky, in 2007, proposed the HEV-M system (Human, Embedded and Virtual agents in Mediation) and published experimental results; However, HEV-M structure, although achieving human-computer interaction, the platform itself and application systems are lack of methods and process as far as software engineering is concerned. Validation of those experiments is also challenged by lack of supports from theories of teamwork.
In this study, based on previous experiences in integrating robots in teaching AI, we first exploit the Robot Team Player and Joint Activity of Woods for interpreting and verifying against the effectiveness of the experiment. We also use an agent-oriented software engineering methodology of Gaia, together with the MSRDS-VPL implementation platform and Lego NXT robots for the Human Robot systems in the pedagogy. Then we apply this approach to the curriculum unit on agent in artificial intelligence. Two related versions of the HEV-M are developed. The HEV-M Replica of the first phase confirms the theoretical framework of this approach, system design methods and analysis and interpretation of behaviors in students participation. The second phase confirms an extension of this method is possible and useful.

摘要 I
Abstract II
謝誌 IV
目錄 V
圖表索引 VIII
1 緒論 1
1.1. 研究背景與動機 1
1.2. 研究目的 4
1.3. 研究架構與流程 4
1.3.1. 研究架構 4
1.3.2. 研究流程 6
1.4. 研究限制 7
1.5. 論文架構 8
2 文獻探討 9
2.1. HEV-M & Bifocal modeling 9
2.2. 團隊玩家(Robot Team Player) & 共同任務理論(Joint Activity) 12
2.3. AOSE-Gaia方法論 14
2.4. 軟體工程實驗性機器人 16
2.5. 模型驅動架構(Model Driven Architecture, MDA) 17
2.6. 三層式機器人 18
2.7. 智慧代理人技術 23
2.8. 機器人開發平台(MSRDS) 24
3 研究方法與步驟 27
3.1. 研究方法說明 27
3.2. AOSE-Gaia方法論分析模型 27
3.2.1. 需求階段 27
3.2.2. 分析階段 28
3.2.3. 設計階段 30
3.2.4. 細節設計階段 36
3.3. 系統架構模型 37
4 AI教案遊戲實驗實作 40
4.1. HEV-M複製實驗(HEV-M Replica Experiment) 40
4.2. 分離擴充實驗(Unbound Extended Experiment) 47
5 資料分析與實驗觀察之驗證 53
5.1. 資料收集、問卷設計與統計方法 53
5.1.1. 資料收集 53
5.1.2. 問卷設計 53
5.1.3. 統計方法 54
5.2. 資料分析 54
5.2.1. Log資料與影像交互分析 54
5.2.2. 開放性問卷資料分析 56
5.2.3. 教學實驗比較 59
6 結論與建議 61
6.1. 結論 61
6.2. 建議 62
參考文獻 63
附錄 66
附錄一、問卷調查表 66
作者簡介 68

[1] 黃申在，潘志傑，鍾凱雯(民96年)，樂高機器人(Lego RCX)應用在人工智慧教學之初探，第十二屆TAAI人工智慧與應用研討會，雲林，雲林科技大學。
[2] 黃申在，潘志傑，鍾凱雯(民97年)，樂高(LEGO)機器人與GreenFoot應用在人工智慧教學之研究，第十三屆TAAI人工智慧與應用研討會，宜蘭，淡江大學
[3] 吳仁和(2005)。物件導向系統分析與設計－結合MDA與UML。台北市：智勝文化。
[4] Morgan, S.(2008). Microsoft Robotics Studio 程式開發 - 新一代超智慧機器人開發平台發(康仕仲、蔡宛庭、涂智超、古凱元)。台北市：悅知文化。
[5] Shen-Tzay Huang, Kai-Wen Chung, Chih-Chieh Pan. (2007), A Case Study in Using LEGO Robots in Introductory Artifical Intelligence, International Workshop on Robotics in Education2007, WRO (World robot Olympiad), Taipei.
[6] Tisue, S. & Wilensky, U. (2004). NetLogo: A simple environment for modeling complexity. International Conference on Complex Systems. Boston MA, May 16-21.
[7] Wilensky, U. (2004). Complexity Perspectives and Multi-agent Modeling in Education. International Conference on Complexity Sciences. Boston, Ma.
[8] Blikstein, P., Rand, W. & Wilensky, U. (2007). Examining Group Behavior and Collaboration using ABM and Robots. In Proc. of Agent .Evanston, IL.
[9] Blikstein, P. & Wilensky, U. (2007). Bifocal modeling: A framework for combining computer modeling, robotics and real-world sensing. At the annual meeting of the American Educational Research Association, Chicago, USA.
[10] Erann Gat, R., Peter Bonnasso, Robin Murphy. (1998). On Three-Layer Architectures. AAAI: Artificial Intelligence and Mobile Robots.
[11] Brugali, D., Agah, A., MacDonald, B., Nesnas, I. & William, S. Smart. (2007). Trends in Robot Software Domain Engineerin. Software Eng. for Experimental Robotics. pp. 3-8.
[12] Russel, S. & Norvig, P. (2003). Artficial Intelligence: A Modern Approach. Upper Saddle River, New Jersey: Pearson Education.
[13] Nostrand, B. (2000). Autonomous Robotics Projects for Learning Software Engineering. Systems, Man, and Cybernetics, 2000 IEEE International Conference on Volume 1, pp. 724 - 729 vol.1.
[14] Zambonelli, F., Jennings, N. & Wooldridge, M. (2003) .Developing Multiagent Systems:The Gaia Methodology. ACM Transactions on Software Engineering and Methodology. Vol. 12, No. 3, pp. 317–370.
[15] Klein, G., P.J. Feltovich, J,M. Bradshaw, and D.D. Woods.(2004).Common ground and coordination in joint activity. In Organizational Simulation, edited by W.B. Rouse and K.R. Boff, 139-184. New York City, NY: Jon Wiley.
[16] Moyaux, T., Chaib, B. & D’Amours, S. (2006). Supply Chain Management and Multiagent Systems: An Overview. Springer Berlin Heidelberg.
[17] Russel, S. & Norving, P. (2003). Artificial Intelligence: A Modern Approach, 2nd eds. Prentice Hall.
[18] Klien, G.; Woods, D.D.; Bradshaw, J.M.; Hoffman, R.R. & Feltovich, P.J. (2004). Ten challenges for making automation a "team player" in joint human-agent activity. IEEE Intelligent Systems. Vol. 19, Issue 6, pp. 91- 95.
[19] Christoffersen, K. & Woods, D.D. (2002). How to make automated systems team players. Advances in Human Performance and Cognitive Engineering Research. Vol. 2, pp.1-12.
[20] Microsoft Robotics Developer Center, http://msdn.microsoft.com/en-us/robotics/default.aspx
[21] Moraitis, P. & Spanoudakis, N. (2006). THE GAIA2JAD PROCESS FOR MULTI-AGENT SYSTEMS DEVELOPMENT. Applied Artificial Intelligence, Vol. 20, pp.251-273.
[22] Castro, A. & Oliveira, E. (2008). The rationale behind the development of an airline operations control centre using Gaia-based methodology. Int. J. Agent-Oriented Software Engineering, Vol. 2, No. 3, pp. 350-377.
[23] Nils J. Nisson. (1980) Principles of Artificial Intelligence. Palo Alto:Tioga.
[24] Rodney A. Brooks. (1986) A Robust Layered Control System for a Mobile Robot. IEEE Journal on Robotics and Automation. Vol. 2, No. 1.
[25] Toal, D., Flanagan, C., Jones, C. & Strunz, B.(1996) Subsumption Architecture for the Control of Robots. University of Limerick.
[26] Issa, A.D. & Nesnas. (2007) The CLARAty Project: Coping with Hardware and Software Heterogeneity. Springer Berlin.