臺灣博碩士論文加值系統

隨著全球人口高齡化現象逐漸昇高之問題，未來個人醫療模式朝向遠距照護及無人看護等技術層面發展。於無所不在運算環境迅速發展下，環境中各異質之自主性感測代理人所得資訊，透過既定競合機制彼此提供資源，將可完成更為複雜之任務。此環境結合遠距式醫療照護及無人看護等技術，可使得未來之個人化醫療達到全天候照護的模式。本論文提出一被照護者行為追踨模型，並運用於非視覺多感測器代理人之自主式智慧照護系統。此行為追踨模型運用因果網路架構，推斷被照護者之行為意圖。並藉由一循序狀態建模技術(Serialized States Modeling)，處理多感測器代理人之異質資訊融合、以及狀態間時序變化等二大行為辨識課題，進而分析出目前被照護者狀態。論文中藉實驗模擬之結果，確認此追踨模型能成功識別出各行為之差異，且隨狀態之時序不同而追踨到正確之行為。因而提昇自主式智慧照護系統於監護被照護者狀態時，識別各特定事件之準確率，及意外事件發生之判別率。

Under modern medical technology, the high-age population is growing up rapidly. In the aging of population circumstances, the aging-people care system is developed and forwards to personalized, autonomous, pervasive, remotely, and daylong. Today, the ubiquitous computing developed rapidly, applies such technology into the high-age people homecare systems is a more important issue. Those systems can perform the complicated homecare task based on the competition to contribution between each heterogeneous and independent agent. In this thesis, a behavior tracking model of home cared aging person is presented and used by our designed intelligent pervasive-care system. This system uses multiple interlaced and non-visual sensors to recognize the behavior of concerned person in independent and concurrent. The behavior tracking model based on the structure of casual network to infer the intention of cared person. Additionally, a serialized states modeling method also developed to solve two major problems about the behavior identification in the aging person care system. Firstly, the fusion of information sensed from the heterogeneous sensor agents in a multiple agent system. Secondly, the behavior discrimination accord to the difference in the sequence of active states. Based on above method, the behavior tracking model can recognize the recent status of the cared person. For confirming our design, a simulation experiment also completed and its results are presented in this thesis, in there the designed model succeed to identify the hidden intention of some specific behavior, and recognize the difference of behaviors when the sequence of active states is changed. Therefore, used this model in our intelligent pervasive-care system can improve the accuracy rate of specific incident, and the discrimination rate about some accidental events.

中文摘要………………………………………………………………………………… i
英文摘要………………………………………………………………………………… ii
誌謝……………………………………………………………………………………… iii
目錄……………………………………………………………………………………… iiii
表目錄…………………………………………………………………………………… v
圖目錄…………………………………………………………………………………… vi
第一章 緒論…………………………………………………………………………… 1
1.1 高齡者之居家照護模型…………………………………………………… 1
1.2 高齡者需求分析…………………………………………………………… 3
1.3智慧型照護系統概述……………………………………………………… 4
1.4 論文概觀…………………………………………………………………… 5
第二章 智慧代理人…………………………………………………………………… 6
2.1 何謂代理人………………………………………………………………… 6
2.2 代理人的分類……………………………………………………………… 8
2.3 物件導向架構與代理人導向架構的不同………………………………… 10
2.4 KQML……………………………………………………………………… 11
2.4.1 KQML傳輸模型……………………………………………………… 11
2.4.2 KQML訊息語法……………………………………………………… 11
2.4.3 Performative參數…………………………………………………… 12
第三章 智慧型普及照護系統………………………………………………………… 14
3.1 系統架構…………………………………………………………………… 14
3.2 多感測器代理人社群……………………………………………………… 15
3.3 監看者子系統……………………………………………………………… 17
3.3.1 評估者………………………………………………………………… 18
3.3.2 協調者………………………………………………………………… 20
3.4 中央協商者………………………………………………………………… 21
3.5 環境模型…………………………………………………………………… 23
3.6 局部行為模型……………………………………………………………… 24
3.7 分散行為模型……………………………………………………………… 26
第四章 被照護者行為追踨模型……………………………………………………… 27
4.1 徵兆層……………………………………………………………………… 29
4.2 事件層、動作層以及行為層……………………………………………… 31
4.2.1 事件層………………………………………………………………… 31
4.2.1 動作層………………………………………………………………… 31
4.2.3 行為層………………………………………………………………… 32
4.3 活動模型…………………………………………………………………… 32
4.3.1 計算行為符合度……………………………………………………… 33
4.3.2 修正行為模型………………………………………………………… 35
4.4 意向層……………………………………………………………………… 36
第五章 實驗結果與分析……………………………………………………………… 38
5.1 調整活動模型……………………………………………………………… 38
5.2 驗證行為識別率…………………………………………………………… 40
5.3 意外事件驗證……………………………………………………………… 44
第六章 結論與未來發展……………………………………………………………… 46
參考文獻………………………………………………………………………………… 47
附錄A…………………………………………………………………………………… 50
簡歷……………………………………………………………………………………… 58

1. 王文俊，認識Fuzzy，全華科技圖書，民國八十六年。
2. 王小川，語音訊號處理，全華科技圖書，民國九十三年。
3. 李世代，「Overview of Elderly Aide」，2005 Embedded Systems Development for Assisted Living Systems and Services Forum，國立台灣大學博理館101室，民國九十四年三月二十二日。
4. 李政益，「特定語者特定中文語音指令雙模辨識技術」，清雲科技大學，碩士論文，民國九十四年六月。
5. 易芝玲，「遠距居家照護服務」，產業科技創新國際研討會，國家圖書館國際會議廳，民國九十二年十月三十一日。
6. 駐美代表處科技組，「未來五年投入新台幣300億元發展六大科技」，駐美代表處科技組，民國九十五年五月八日。
7. 鄭智源，徐業良，「建立高齡者生活品質之智慧型維護系統」， 工業工程雜誌，民國九十一年十二月。
8. 鄭智源，徐業良，「居家環境下健康監測技術相關研究介紹」，元智大學老人福祉科技研究中心，民國九十年二月十九日。
9. 鄭智源，「自動化居家健康監測系統之開發」，元智大學老師福祉科技研究中心，民國九十年六月十二日。
10. 蘇木春，張孝德，機器學習：類神經網路、模糊系統以及基因演算法則，全華科技圖書，民國八十八年。
11.Akifumi Yamaguchi, Mitsuhiro Ogawa,　Toshiyo Tamura, “Monitoring behavior in the home using positioning sensors”, IEEE, Medicine and Biology Society, Volume 20, Issue 4, pp. 1977-1979, 1998,
12. Carol A. Miller, “Nursing Care of Older Adults: Theory and Practice”, Williams & Wilkins, 1995.
13. Chin-Teng Lin, C.S. George Lee, “Neural Fuzzy Systems: A Neuro-Fuzzy Synergism to Intelligent Systems”, Prentice Hall, 2003.
14. Chao-Lin Wu, Wei-Chen Wang, Li-Chen Fu, “Mobile Agent Based Integrated Control Architecture for Home Automation System”, IEEE, International Conferences on Intelligent Robot Systems , 2004.
15.Elizabeth D. Mynatt, Anne-Sophie, Melenhorst, Arthur D. Fisk, and Wendy A. Rogers, “Aware Technologies for Aging in Place: Understanding User Needs and Attitudes”, IEEE, Pervasive Computing, Volume 03, Issue 2, pp.36-41, April-June, 2004
16. Hani Hagras, Victor Callaghan, Martin Colley, Graham Clarke, “Creating an Ambient-Intelligence Environment Using Embedded Agents”, IEEE, INTELLIGENT SYSTEMS, Volume 19, Issue 6, pp. 12-20, Nov-Dec 2004.
17. Hanumant Deshmukh, Jignesh Malavia, Matthew Scarpino, SCWCD Exam Study KIT Second Edition: Java Web Component Developer Certification, DrMaster Press, 2005.
18.Jay Weber , Tim Finin, Gio Wiederhold, “DRAFT Specification of the KQML Agent-Communication Language”, June 1993.
19.John Yen and Reza Langari, Fuzzy Logic: Intelligence, Control, And Information, Prentice Hall, 1998.
20.Jacques Ferber, Multi-Agent Systems: An Introduction to Distributed Artificial 　　　　　　　　　　　　　　　Intelligence, Addison Wesley, 1999.
21. Mark Perry, Alan Dowdall, Lorna Lines, “Multimodal and Ubiquitous Computing Systems: Supporting Independent-Living Older Users”, IEEE, Transactions on Information Technology in Biomedicine, Vol. 8, No. 3, pp.258-270, September 2004.
22. Matthai Philipose, et al, “Inferring Activities from nteractions with Objects”, IEEE, Pervasive Computing, Volume 03, Issue 4, pp.50-57, October-December, 2004.
23.Mitsuhiro Ogawa, et al, “An attempt of monitoring daily activities at home”, IEEE, EMBS International Conference, pp. 786-788, July 2000.
24. Stuart Russell, Petter Norvig, Artificial Intelligence: A Modern Approach, Prentice Hall, Upper Saddle River, 1995.
25.Thomas Schetter, Mark Campbell, Derek Surka, “Multiple Agent-based Autonomy for Satellite Constellations”, Artificial Intelligence, pp. 147-180, 2003.
26. Tim Finin, “Specification of the KQML Agent Communication Language”, Technical report, DARPA Knowledge Sharing Initiative, External WorkingGroup, 1992.
27. Tim Finin, “DRAFT Specification of KQML Agent-Communication Language. Plus example agent policies and architectures”, The DARPA Knowledge Sharing Initiative External Interfaces Working Group , 1993.
28.Yannis Labrou and Tim Finin, “A Proposal for a new KQML Specification” , February 3, 1997.