基於本體論的情境感知大樓自動化系統

An Ontology-based Context-Aware Building Automation System

周金旺(Jin-Wang Zhou)

指導教授 ： 段裘慶

共同指導教授 ： 蕭榮修(Rong-Shue Hsiao)

國立臺北科技大學/電資學院/電腦與通訊研究所/碩士(2014年)

https://doi.org/10.6841/NTUT.2014.00589

摘要

現代化的大樓自動化系統(building automation system, BAS)可提供情境的服務(situated service)，即系統可隨情境的變化提供使用者適當的服務，此系統謂之情境感知(context-aware)系統。隨著無線感測網路的蓬勃發展，並具有低佈建成本、低功耗和高擴充性等優點，情境感知系統可透過無線感測節點的密集佈建，達到精細的控制。本論文提出一個基於本體論的情境感知系統架構，並實作一個情境感知的大樓自動化系統，包含感測資料融合、本體論情境模型的建立及情境推理等技術。
情境模型是建構情境感知系統的基礎，情境模型包含感測器、控制的設備和空間等情境資料。而本體語言為描述詞彙之間關係的一種語言，可以充分表達出實體的概念及與實體之間的關係，使用本體語言來建構情境模型，不僅可以模擬情境模型中各種情境資料的實體概念，也可驗證情境系統的可行性及完整性，本論文使用Protege軟體來建構本體論情境模型。然而，人員在場的情境資料是情境感知系統中最為關鍵的情境資料，本論文提出一個低計算複雜度的多感測資料融合演算法，並與基於動態貝式網路的多感測資料融合演算法比較。實驗結果顯示，本論文所提出的演算法和基於動態貝式網路的資料融合技術之準確度分別達96.96%和96.67%的高準確率。實作的結果亦顯示，本論文提出的架構可以有效地實現於情境感知的大樓自動化系統，此架構可適用於其他的情境感知系統之建構。

關鍵字

情境感知 ； 本體論 ； 資料融合 ； 在場情境資料

並列摘要

Modern building automation system could provide situated service. According to the different context, the system could provide appropriate service to the user, the so called a context-aware system. As the technology of wireless sensor network becomes more mature, this provides features of low cost, high scalability, and energy efficiency. A context-aware system can achieve fine control through densely deployed sensor nodes. This paper proposed an ontology-based context-aware system architecture and implemented a context-aware building automation system which includes sensory data fusion, ontology context model and context inference.
Context model is a basis to establish context-aware system, including the relationship between context data, the presents of context data and its applications. The ontology language can be used to describe the relation between vocabularies. It could present the property of entity and relation with others. Using ontology language to establish context model not only simulates the physical concepts of various context data in context model, but also verifies the feasibility and completeness of context-aware systems. This paper used Protege platform to establish ontology context model. Occupancy data is the most important context data in a context-aware system. The proposed multimodal sensor fusion algorithm with low complexity was compared with the dynamic Bayesian network-based multisensor data fusion algorithm. The experimental results showed that the proposed multimodal sensor fusion algorithm and the dynamic Bayesian network-based multisensor data fusion accuracy are 96.96% and 96.67%, respectively. Also, it is shown that the proposed architecture could be applied efficiently in context-aware building automation system and other context-aware systems.

並列關鍵字

Context-aware ； Ontology ； Data fusion ； Occupancy context

參考文獻

[1] M. Weiser, “The computer for the 21st century,” Scientific American, vol. 265, no. 3, 1991, pp. 94-10.

[2] I. F. Akyildiz, W. Su, Y. Sankarasubramaniam and E. Cayirci, “Wireless Sensor Networks: A Survey,” Computer Networks, vol. 38, no. 4, 2002, pp. 393-422.

[3] A. K. Dey, D. Salber and G. D. Abowd, “A Conceptual Framework and a Toolkit for Supporting the Rapid Prototyping of Context-Aware Applications,” Human-Computer Interaction, vol. 16, no. 2, 2001, pp. 97-166.

[7] S. Lee, K. N. Ha and K. C. Lee, “A Pyroelectric Infrared Sensor-based Indoor Location-Aware System for the Smart Home,” IEEE Transactions on Consumer Electronics, vol. 52, no. 4, 2006, pp. 1311-1317.

[8] K. N. Ha, K. C. Lee and S. Lee, “Development of PIR Sensor based Indoor Location Detection System for Smart Home,” Proceedings of the International Joint Conference on Society of Instrument and Control Engineers and Institute of Control, Automation and System Engineers, 2006, pp. 2162-2167.