隨著物聯網的發展跟服務型機器人的普及，在智慧家庭中的管理者可透過感應器所測量到的資訊來判斷居民的行為並操控不同功能的服務型機器人，去給予居民需要的服務。不同種類跟功能的機器人有可能是透過不同軟體驅動，因此要讓管理者操作異質的機器人變得不容易。另一方面，感應器會因環境限制或本身故障，無法一直提供正確的感測結果，這會導致不確定性，造成管理者無法正確辨識居民的行為。本研究利用基於OSGi框架的OneM2M (OM2M)中介軟體平台，來解決異質機器人管理問題並提供可靠的行為辨識。我們提出的架構主要分為三層，分別是Client Layer、Server Layer跟Robot/Sensor Layer。Client Layer透過Server Layer，能跟Robot/Sensor Layer溝通。管理者在Client Layer透過瀏覽器使用我們在Server Layer所提供的圖形化介面，在同一個網頁能夠容易操作異質的機器人。在Server Layer內，我們使用OM2M中介軟體提供各式資源管理服務給不同類型的機器人，並且來接收來自Client對異質機器人的請求，再轉傳至其對應的機器人資源。最後在Robot/Sensor Layer中，我們實作多種類型的機器人和可上網的感應器，將感測結果和機器人狀態，經由中介軟體的翻譯，傳送至Server Layer，以提供Client Layer的存取。我們也在Server Layer加入基於Dempster-Shafer證據理論(DS Theory)的服務包，此服務包考量感應器的結果的不確定性，產出信任值及不確定值，供管理者更加可靠地辨識居民的行為。基於OSGi的框架下開發的服務包，能因應異質機器人的加入移除或感測環境的不同，無需重新啟動系統即可被遠程安裝、啟動和卸載，達到一個具彈性的架構。最後我們透過探討不同的感應器佈署對於辨識結果的影響，可以在一定的準確性要求下，啟動較少的感測器，達到節能效果。

With the development of the Internet of Things and the popularity of service robots, in smart home, managers can recognize the inhabitant’s activity and control distinct types of robots through the sensing result obtained by Internet-enabled sensors to provide different services to inhabitants. Different types of service robots might be driven by different software platforms, so it is not easy for managers to control heterogeneous robots. In addition, the environment limitation or the failure of sensors results in the wrong sensing result, misleading the wrong manager’s decision. This study uses OnM2M (OM2M) based on OSGi framework as the middleware platform to solve the problem of heterogeneous service-robot management while providing the dependable activity recognition. The proposed architecture contains three layers, namely Client Layer, Server Layer and Robot/Sensor Layer. Managers can control heterogeneous robots on the same webpage through the graphical user interface provided in the Client Layer. In Server Layer, We use OM2M middleware to provide various resource management services to distinct types of robots, and receive requests from clients for heterogeneous robots, and then transfer to their corresponding robot resources In Robot/Sensor Layer. Finally, in the Robot/Sensor Layer, we implement various types of robots and internet-accessible sensors, and transmit the sensing results and Robot state to the Server Layer, then user can obtain information in Client Layer through the translation of the middleware. In addition, we have provided the Dempster-Shafer theory (DS Theory) service bundle in the Server Layer that produces the belief value and uncertainty value based on the fusion result of sensors to help managers recognize the activity dependably. Based on OSGi framework, we can remotely install, active or uninstall these service bundles according to the addition or the removal of robots or the modified environment sensors without rebooting the whole system, which achieves a flexible architecture. Finally, the proposed method can activate the fewer number of sensors and achieve energy saving within a certain of dependability based on the evaluation of the effect for the number of sensors on the sensing results.

[1] A. Al-Fuqaha, M. Guizani, M. Mohammadi, M. Aledhari, and M. Ayyash, "Internet of Things: A Survey on Enabling Technologies, Protocols, and Applications," IEEE Communications Surveys & Tutorials, vol. 17, no. 4, pp. 2347-2376, 2015.
[2] M. Naphade, G. Banavar, C. Harrison, J. Paraszczak, and R. Morris, "Smarter Cities and Their Innovation Challenges," Computer, vol. 44, pp. 32-39, 2011.
[3] Pepper. "Pepper the humanoid robot", [Online]. Available: https://www.softbankrobotics.com/emea/en/pepper (accessed October 25, 2018).
[4] Zenbo. "Zenbo - Your Smart Little Companion". [Online]. Available: https://zenbo.asus.com/tw/product/overview (accessed October 25, 2018).
[5] ROS. "Robot Operating System", [Online]. Available: http://www.ros.org/ (accessed December 6, 2018).
[6] J.-B. Yang, D.-L. Xu, "Evidential reasoning rule for evidence combination", Artif. Intell., vol. 205, pp. 1-29, Dec. 2013.
[7] G. Shafer, "A Mathematical Theory of Evidence", Princeton University Press, 1976.
[8] ETSI. [Online]. Available: http://www.etsi.org/ (accessed October 25, 2018).
[9] OneM2M. [Online]. Available: http://www.OneM2M.org/ (accessed October 25, 2018).
[10] OneM2M-TS-0001, "Functional Architecture", v2.10.0, [Online]. Available: http://www.OneM2M.org/images/files/deliverables/Release2/TS-0001-%20Functional_Architecture-V2_10_0.pdf (accessed October 25, 2018).
[11] OSGi Alliance. [Online]. Available http://www.osgi.org/ (accessed October 25, 2018).
[12] M. Quigley, K. Conley, B. Gerkey, J. Faust, T. Foote, J. Leibs, R. Wheeler, A. Y. Ng, "ROS: an open-source Robot Operating System", ICRA workshop on open source software, vol. 3, no. 3.2, pp. 5, 2009.
[13] EV3. [Online]. Available: https://www.lego.com/en-us/mindstorms/about-ev3 (accessed October 25, 2018).
[14] LeJOS. [Online]. Available: http://www.lejos.org/ (accessed October 25, 2018).
[15] ev3dev. [Online]. Available: https://www.ev3dev.org/ (accessed October 25, 2018).
[16] hacks4ros. [Online]. Available: http://hacks4ros.github.io/h4r_ev3_ctrl/ (accessed October 25, 2018).
[17] R. Esteller-Curto, E. Cervera, A. P. del Pobil, R. Marin, "Proposal of a rest-based architecture server to control a robot", Innovative Mobile and Internet Services in Ubiquitous Computing (IMIS) 2012 Sixth International Conference on, pp. 708-710, July 2012.
[18] REST. "Rest architectural style". [Online]. Available: http://c2.com/cgi/wiki?RestArchitecturalStyle. (accessed October 25, 2018).
[19] G. Li, H. Wang, X. Ying, J. Liu, "A proxy-based cloud infrastructure for home service robots", The 27th Chinese Control and Decision Conference (2015 CCDC), pp. 5718-5723, May 2015.
[20] L. F. Costa, L. M. G. Goncalves, "Roboserv: A ROS based approach towards providing heterogeneous robots as a service", 2016 XIII Latin American Robotics Symposium and IV Brazilian Robotics Symposium (LARS/SBR), pp. 169-174, Oct 2016.
[21] L. Chen, J. Hoey, C. Nugent, D. Cook, Z. Hu, "Sensor-based activity recognition", IEEE Trans. Syst. Man Cybern. C Appl. Rev., vol. 42, no. 6, pp. 790-808, Nov. 2012.
[22] L. Chen, C. Nugent, G. Okeyo, "An ontology-based hybrid approach to activity modeling for smart homes", IEEE Trans. Human-Mach. Syst., vol. 44, no. 1, pp. 92-105, Feb. 2014.
[23] J. Bobadilla, F. Ortega, A. Hernando, J. Bernal, "A collaborative filtering approach to mitigate the new user cold start problem", Knowl.-Based Syst., vol. 26, pp. 225-238, Feb. 2012.
[24] Ghasemi, V., A. Pouyan, and M. Sharifi. "A Sensor-Based Scheme for Activity Recognition in Smart Homes using Dempster-Shafer Theory of Evidence." Journal of AI and Data Mining 5.2 (2017): 245-258.
[25] X. Hong, C. Nugent, M. Mulvenna, S. McClean, B. Scotney, and S. Devlin, “Evidential fusion of sensor data for activity recognition in smart homes,” PMC, vol. 5, no. 3, pp. 236–252, June. 2009.
[26] ROS Bridge. [Online]. Available: http://wiki.ros.org/rosbridge_suite (accessed October 25, 2018)
[27] D3.js. [Online]. Available: https://d3js.org/ (accessed October 25, 2018).