應用階層式行為建議與回復機制於智慧家庭

Application of Hierarchical Behavior Suggestion and Recovery Mechanism to Smart Homes

陳建宏(Chieh-Hung Chen)

指導教授 ： 沈榮麟

國立臺北大學/電機資訊學院/電機工程研究所/碩士(2012年)

摘要

本研究提出以階層式觀念應用到數位家庭(Smart Homes)代理人管理平台中的行為建議系統與回復機制。數位家庭可以分成區域(Location)層、動作層(Action)、家電設備(Home Appliance)層等；區域管理動作，動作管理家電設備的階層式架構管理。本論文提出Hierarchical Human Behavior Suggestion Algorithm(HHBSA)演算法建議使用者的行為模式，演算法分為Location-Learning Suggestion Algorithm(LLSA)與Action-Behavior Suggestion Algorithm(ABSA)，LLSA利用Q-learning與Fuzzy-State Q-learning(FSQL)的概念來建議使用者的使用區域。ABSA根據使用者所在區域修正後的建議區域序列，建議區域內的建議行為。建議行為後可以預先啟動行為所包含的家電設備，當設備啟動發生錯誤時，以階層式回復機制來試圖修復錯誤。將行為層設為回復點，設備錯誤時，重新執行行為內的設備。回復點可以根據使用順序而改變，動態的改變回復點使行為系統可以無限增加行為且維持回復機制的效率。

關鍵字

智慧家庭 ； Q-learning Algorithm ； 階層式架構 ； 回復機制

並列摘要

In this thesis, we proposed the behavior suggestion system and recovery mechanism applied to the smart home management platform with a hierarchical structure; the smart home system can be divided into location layer, action layer, and home appliance layer. The smart home management system uses the hierarchical structure to reach regional management action and action management appliances. This study also provides Hierarchical Human Behavior Suggestion Algorithm (HHBSA), and suggests the behavior pattern. HHBSA includes Location-Learning Suggestion Algorithm (LLSA) and Action-Behavior Suggestion Algorithm (ABSA); LLSA suggests usage location of user with the concepts of Q-learning and Fuzzy-State Q-learning (FSQL). ABSA provides advice on regional behaviors according to the suggested regional sequence updated in user’s location. The home appliances included in behaviors can be switched on in advance when suggested behaviors have been provided. A hierarchical recovery mechanism may be used to revise errors occurred while starting the home appliances. The home appliances can be re-executed when errors occurred if the action layer is set as a recovery point that can be changed according to using sequence. A dynamic recovery point makes it possible to unlimitedly add behaviors for a behavior system, and to maintain the efficiency of recovery mechanism.

並列關鍵字

Smart Home ； Q-learning Algorithm ； Hierarchical Structure ； Recovery Mechanism

參考文獻

[1] Sunha Bea, S. W. Lee, Y. S. Kim, and Z.Bien, "Fuzzy-state Q-learning-based human behavior suggestion system in intelligent sweet home," Proc. Int. Conf. Fuzzy Systems, pp. 283-287, 2009.

[2] Lu-Yu Chen, " A SOM-based fuzzy systems Q-learning in continuous state and action space," Master thesis, Univ. of National Central , 2006.

[3] Zizhong Chen and Dongarra J., “Algorithm-based fault tolerance for tail-stop failures,”, IEEE Trans. Parallel and Distributed Systems , vol.19, pp.1628-1641, 2008.

[4] Ge-Ming Chiu and Jane-Ferng Chiu, “A new diskless checkpointing approach for multiple processor failures,” IEEE Trans. Dependable and Secure Computing, vol.8, pp.481-493, 2011.

[5] M. A. Feki, S. W. Lee, M. Mokhtari and Z. Bien, “Context aware life pattern prediction using fuzzy-state Q-Learning,” Proc. Int. Conf. Smart home and Telecommunication, pp.188-195 , 2007 .