Knowledge Discovery in 3D Geospatial Spaces: Towards an Integrated BIM, 3D GIS and IoT Data Framework

世界正在进入智慧城市时代。物联网(IoT)和信息技术基础设施是智能城市的基础，智能城市是由互联的智能设备、传感器和大数据分析组成的网络。据估计，到2020年，全球将有800亿台联网设备和超过50亿互联网用户。连接是不可避免的，它将包括我们生活的方方面面，无缝地将工作、家庭和周围环境连接在一起。在这种背景下，3D虚拟城市空间是帮助智慧城市管理和监测室内外动态的关键。加拿大城市决心实现成为智慧城市的目标。为了促进动态的城市应用，如业务增长和监控基础设施的解决方案，多伦多和蒙特利尔都向公众推出了3D开放数据。然而，需要进一步的研究来处理虚拟空间中的多尺度分析和可视化，从城市尺度延伸到建筑结构，反之亦然。 这项研究的主要目标是增加我们对来自建筑信息模型(BIM)、3D地理信息系统(GIS)和物联网的各种数据如何集成到单一环境中进行时空分析的了解。我们的研究将通过多维图形结构识别集成数据的本体、语义和几何定义来促进知识的发展。9名HQP将通过以下工作进行培训：1)处理BIM、地理信息系统和物联网数据流的全面、集成的数据建模框架；2)3D室内外虚拟模型中数据挖掘的分析方法，以发现知识；3)在无缝的室内外虚拟模型中多尺度表示已发现的知识的空间聚合操作；4)用于BIM、地理信息系统和物联网集成的地理空间数据共享、互操作性和质量评估框架。 我们从研究中获得的知识将使新技术能够通过与物联网流信息连接的虚拟空间来聚合和细化海量3D室内外数据的图像。这一知识将对加拿大地理空间工业部门产生重大影响。我们的研究还将为3D数据共享的标准化提供技术进步、创新方法和深刻见解。该成果将应用于智能城市的广泛应用，如能源效率、应急管理、基础设施监测和设施管理。

The world is entering the period of smart cities. The Internet of Things (IoT) and information-technology infrastructure are fundamental to smart cities, a network of connected smart devices, sensors, and big-data analytics. By 2020, estimates are that there will be 80 billion connected devices worldwide and over 5 billion Internet users. Connectivity is inevitable and will include all facets of our lives, seamlessly bringing together work, home and surrounding environments. In this context, 3D virtual city space is key to helping smart cities manage and monitor indoor-outdoor dynamics. Canadian cities are determined to achieve the goal of becoming smart cities. To foster dynamic city applications such as business growth and solutions for monitoring infrastructure, Toronto and Montreal have both launched 3D open data to the public. However, further research is required to deal with multi-scale analysis and visualization in a virtual space, extended from city-scale to building structures and vice-versa seamlessly through the temporal scale. The main goal of this research is to increase our knowledge of how a variety of data coming from Building Information Model (BIM), 3D Geographical Information Systems (GIS) and IoT, integrate into a single environment to conduct spatiotemporal analytics. Our research will advance knowledge by identifying ontology, semantics and the geometry definition of integrated data through a multidimensional graph-structure. Nine HQP will be trained by working on the development of 1) A comprehensive, integrated data modeling framework handling BIM, GIS and IoT stream data; 2) An analytical approach for data mining in 3D indoor-outdoor virtual model in order to discover knowledge; 3) Spatial aggregation operations for multi-scale representation of discovered knowledge in a seamless indoor-outdoor virtual model; 4) A geospatial data sharing, interoperability and quality assessment framework for the integration of BIM, GIS and IoT. Knowledge gained from our research will enable new technologies that have an aggregated and granulated image of massive amounts of 3D indoor-outdoor data through a virtual space connected with IoT stream information. This knowledge will have a significant impact on the Canadian geospatial industry sectors. Our research will also provide technological advancement, innovative approaches and deep insight into the standardization of 3D data sharing. The results will be applied to a wide range of smart cities applications such as energy efficiency, emergency management, infrastructure monitoring, and facility management.