Smart indoor Spaces (SiS): Towards an Integrated Framework for the Internet of Things in the Built Environment

物联网（IoT）技术的进步有望彻底改变我们生活的各个方面，其中关键是我们生活和工作的空间。嵌入建筑环境中的传感器可以系统地观察和量化我们的活动。拟议的研究计划将建筑物视为网络物理系统，并整合了两条研究线，到目前为止，这两条研究线一直被单独对待：（a）智能建筑研究，旨在改善大型建筑物的能源消耗，（B）智能家居研究，旨在为体弱者的日常生活活动提供个性化的支持，提高他们的安全性和舒适性。这两条研究线之间的这种人为分离，即使两者都旨在提取关于建筑物居住者活动的信息，尽管在不同的粒度级别，也从根本上限制了它们的算法贡献和它们的工具的采用的一般性。 基于我们最近在智能公寓和BIM建模项目上的工作，拟议的研究计划将制定一个连贯的框架来建模智能室内空间（它们的物理结构，它们包含的传感器，执行器和设备及其行为，以及它们的居住者的活动），以改善建筑物的管理和居住者的舒适度。拟议的研究是围绕三个高层次的主题。“模型和核心算法”主题将提供一个丰富而连贯的室内空间模型，以实现对空间属性（美学，人体工程学，可访问性等）的推理，这些空间中替代传感器部署的真实模拟，以及基于传感器的应用程序的微调，其多模态数据流的一致同步。“用户交互”主题将开发和评估方法，以帮助用户理解传感器数据和通过传感器融合推断的信息，并控制其环境中的系统。它还将为居住者提供一个用户交互框架，以传达他们的偏好，并相应地控制SiS平台。“访问控制”主题将为管理对SiS组件的访问产生可用、可审计和安全的模型，即，物理空间和硬件，以及传感器发出的数据和信息。通过这些活动开发的方法和工具将集成到一个可配置的面向服务的平台中，我们将在两个重要的现实用例的背景下进行评估：家庭脆弱性评估和商业建筑能耗的减少。这项评估工作将与合作伙伴组织合作进行，这些组织将从SiS应用程序的部署中受益，即格伦罗斯康复医院和Covenant Health（用于虚弱评估）以及埃德蒙顿市通过我们最近建立的MetroLab合作伙伴关系（用于能源管理）。

The advancement of Internet of Things (IoT) technologies promises to revolutionize all aspects of our life, and key among them are the spaces where we live and work. Sensors embedded in the built environment enable the systematic observation and quantification of our activities. The proposed research program views buildings as cyberphysical systems and integrates two lines of research that, up until now, have been treated separately: (a) Smart Buildings research, aiming to improve the energy consumption of large buildings, and (b) Smart Homes research, aiming to provide personalized support to frail individuals in their activities of daily living, increasing their safety and comfort. This artificial separation between these two lines of research, even when both aim to extract information about the activities of the buildings' occupants albeit at different granularity levels, fundamentally limits the generality of their algorithmic contributions and the adoption of their tools. Grounded on our recent work on the Smart Condo and BIM modeling projects, the proposed research program will formulate a coherent framework for modeling Smart indoor Spaces (their physical structure, the sensors, actuators and devices they contain and their behaviors, and the activities of their occupants) in order to improve the management of buildings and the comfort of their occupants. The proposed research is organized around three high-level themes. The "Models and Core Algorithms" theme will deliver a rich and coherent model of indoor spaces to enable reasoning about the properties of the space (aesthetics, ergonomics, accessibility, etc.), realistic simulations of alternative sensor deployments in these spaces, and fine-tuned sensor-based applications with consistent synchronization of their multimodal data streams. The "User Interaction" theme will develop and evaluate methods to help users understand the sensor data and the information inferred through sensor fusion, and to control the systems in their environment. It will also produce a user-interaction framework for occupants to convey their preferences and accordingly control the SiS platform. The "Access Control" theme will produce usable, auditable and safe model for managing access to the SiS components, i.e., the physical spaces and hardware, as well as the sensor-emitted data and information. The methods and tools developed through these activities will be integrated in a configurable service-oriented platform, which we will evaluate in the context of two important real-world use cases: at-home frailty assessment and reduction of energy-consumption in commercial buildings. This evaluative work will be conducted in collaboration with partner organizations who stand to benefit from the deployment of SiS applications, namely the Glenrose Rehabilitation Hospital and Covenant Health (for frailty assessment) and the City of Edmonton through our recently established MetroLab partnership (for energy management).