SMART Cities: The need for spatial intelligence

For the first time in human history, more than half of the global population lives in urban areas. This will increase to 70% by 2050. Shanghai’s population has almost doubled in a decade, from less than 13 million residents in 2000 to an estimated 23 million today, and by 2050 it is expected to exceed 50 million. Cities cover just 2% of the Earth’s surface yet consume about 75% of the world’s resources. So it becomes obvious that cities are the key element when coping with climate change and reduction in the use of resources. Since city growth can hardly be avoided, one must be able to cope with its consequences. Here it is essential that harmony exists or is generated among the spatial, social, economical and environmental aspects of a city and between their inhabitants. This harmony hinges on three key pillars: Earth environment, economic development and social equity. These pillars are balanced through sustainability. In this context the concept of a SMART city has emerged. Usually, “smartness” is expressed by its 6-axes model: smart economy, mobility, environment, people, living and governance. Only if all these elements are in balance a city can fulfil its request for sustainability and quality of life. In other words, a city can be called “smart” if investments in human and social capital and traditional (transport) and modern information and communication technologies, information and communication infrastructure will fuel sustainable economic development, a high quality of life, with a wise management of natural resources, through participatory governance.

A smart city possesses Spatial intelligence. This summarizes all components in terms of brain-, hard- and software, which are required to manage a city efficiently with the goal to sustain high quality of life over a long period of time (resilience). As such, it refers to informational and cognitive processes, such as information collection and processing, real-time alert, forecasting, learning, collective intelligence and distributed problem solving. In this environment, the Geo-Spatial Information Sciences play a key role, providing for the underlying theoretical framework and practical procedures for data acquisition, processing, analysis and representation.

Traditional methods of planning and managing cities do not work anymore in this new environment. New approaches are necessary. While interactive design and planning will still be of need and value, computerized techniques must find more interest. It is generally agreed that context is a key element in the design of future, new and the transformation of existing cities. Context must include societal, governmental, economical and technological components. This context must be transformed to and modelled in the digital (computer) domain. The so-created models must be supported and activated by data. The complexity of the political, social and economic decision-making processes requires precise, reliable, actual and largely complete data. Most of this data are spatially related. This clearly emphasizes the relevance of the Spatial Information Sciences for any future-oriented design and planning process.

In Schmitt (1) the “Digital Chain” is promoted, which is a concept of data acquisition, data handling and decision-making on an architectural scale, involving design, construction and facility management of buildings. This “digital chain” is reflected in the concept and work of the SEC-Future Cities Laboratory. The Singapore-ETH Centre (SEC) for Global Environmental Sustainability was established by ETH Zurich and Singapore’s National Research Foundation (NRF) in 2010. It is an institution that frames a number of research programmes, the first of which is the Future Cities Laboratory (FCL). The SEC strengthens the capacity of Singapore and Switzerland to research, understand and actively respond to the challenges of global environmental sustainability. It is motivated by an aspiration to realize the highest potentials for the present and future societies. SEC serves as an intellectual hub for research, scholarship, entrepreneurship, postgraduate and postdoctoral training. It actively collaborates with local universities and research institutes and engages researchers with industry and government agencies to facilitate technology transfer for the benefit of the public. The FCL is a highly trans-disciplinary research centre focused on urban sustainability in a global frame. It is home to a community of over 100 PhD, postdoctoral and Professorial researchers working on diverse themes related to future cities and environmental sustainability. For details, see http://futurecities.ethz.ch.Within this project, a city is seen as an urban metabolism, and the concept of stocks and flows is used to describe and analyse its status and dynamics. This is done on three different levels of scale: S (small)-scale: the individual building, M (medium)-scale: the urban part and L (large)-scale: the territory. Ten different research modules investigate into stocks and flows of energy, materials, capital, people, water, space and information. The information part is treated on the Simulation Platform. Its goal is to support design and decision-making processes with new techniques and approaches to data acquisition and processing, information visualization and simulation. The Simulation Platform currently encompasses more than 20 researchers. Some of them contribute as authors to this volume.

Why is FCL stationed in Singapore and why is in general research and development in the many fields contributing to the concept of Smart City highly developed and acknowledged in Singapore? As the competition for resources increases and urban populations expand, Singapore is an interesting example for a city on its way to mega-structures, which goes a long way in order to cope with the problems of achieving sustainable development and resilience on the long run. Many measures of the past and currently combine to make Singapore a smarter city. “What we have done is to research and try to distill the principles for Singapore’s success in sustainable urban development – we call it a liveability framework”, says Khoo Teng Chye, executive director at the Centre for Liveable Cities based in Singapore.

The “digital chain” may also serve as a synonym for the role that Geomatics plays in the context of Smart Cities. If we understand Geomatics as the science of acquiring, modelling, analysing and representing spatially referenced data, then it integrates as key disciplines Geodesy, Geodetic Mensuration, Photogrammetry and Remote Sensing, Cartography and Geoinformatics. Much of the work of the SEC-FCL Simulation Platform is concerned with Geomatics issues. Some of the Geomatics-related R & D topics of the Simulation Platform are

•	Automatic or semi-automated generation of Digital Surface Models (DSM) from satellite, aerial and terrestrial images and/or LiDAR data,
•	Further development of the semi-automated techniques (like CyberCity Modeler) onto a higher level of automation,
•	Integrated automated and semi-automated processing of laser-scan point clouds and images, both from aerial and from terrestrial platforms,
•	Streamlining the processing pipeline for UAV image data projects,
•	Exploring the various applications of UAV-based thermal imaging,
•	Set-up of GIS with 3D/4D capabilities,
•	Change detection and updating of databases,
•	Combination of real and synthetic (e.g. planned) objects (reality-based and generic modelling) - see CC-Modeller and City Engine,
•	Handling of dynamic and semantic aspects of city modelling and simulation. This leads to 4D city models,
•	LBS system investigations (PDAs, mobiles),
•	Establishment of a powerful visualization and interaction platform (“Value Lab Asia”).

In Geomatics R&D, as well as in professional practice, five major forces are currently at work:

•	New sensors – from optical to microwaves, from satellite to close-range platforms. Multisensor arrangements are becoming more popular and powerful,
•	Processing power – computers and software are becoming more efficient,
•	Data storage – huge amounts of data can be stored at relatively low cost,
•	Connectivity – Internet and cloud computing allows for new ways of information transfer and collaboration,
•	Visualization, simulation and interaction – Communication and insight into the nature of data and processes are a lot easier to gain.

Besides its own R & D programme the Simulation Platform has a distinct service function with respect to all other SEC-FCL scientists. A major component of this service function is the maintenance of a GIS and the training of other researchers in GIS-related topics.

Currently, the GIS database includes data from Singapore (and partly from Indonesia and Malaysia) like:

•	DTMs, DSMs at different resolutions,
•	Topographic maps,
•	Master plans in vector format,
•	Land use maps, drainage patterns,
•	Buildings: coordinates, building types, number of floors above and below ground, number of flats and rooms, roof type and shape, type of ownership value (insured value/market value), status of protection as heritage, life cycle of the lot/buildings, age,
•	Navteq navigation map,
•	Historical plans/cadastre maps,
•	Census data 2010 with location,
•	Georeferenced postcodes,
•	Climate/weather data, temperature of ground at various depths, annual temperatures of rivers and ocean,
•	40 layers of POIs, Singapore address points, Singapore detailed control plans,
•	3D city models of Rochor, Punggol and NUS campus

Among all the Geomatics technologies, we can address in this issue only a few. The team of authors has been carefully selected, making sure that crucial techniques are properly presented. With the topic “Geomatics for Smart Cities” being so broad, we can only show a small segment of all the technologies involved in present-day activities.

Gerhard Schmitt opens the contributions with “Spatial Modelling Issues in Future Smart Cities”. He looks at the role of geometry and space in the continuum of city modelling and briefly outlines the development from an originally “rule-based” approach, through the “stocks and flows” concept, to the current “complex system and quantum city” approach. Then, he describes more in detail the “stocks and flows” model, as it is used in the current work of the SEC-FCL team, addressing material, energy, water, people, finances, density and space and information. The stocks and flows of information are handled on the Simulation Platform. It is clearly stated that although the role of geometry and spatial issues have changed over the centuries, they continue to be foundations for urban design decisions in a Smart City context.

At a finer level of technical detail, Li Deren, Shan Jie, Shao Zhengfeng, Zhou Xiran and Yao Yuan are presenting an overview contribution dealing with “Geomatics for Smart Cities – Concept, Key Techniques and Applications”. They outline the current shift from Digital Cities to Smart Cities. They see Digital Cities, The Internet of Things and Cloud Computing as important supporting technologies for Smart Cities and discuss how these technologies can be implemented and then will contribute to a better management of cities. As new technologies for Smart Cities, they describe briefly the location cloud, remote sensing cloud, integration of video and GIS, integration of space-borne, air-borne and terrestrial sensors and GIS, indoor and underground navigation, ubiquitous sensing via smart phones and spatio-temporal data mining. Finally, they show four concrete “smart” applications in Wuhan and Sichuan: Smart municipal supervision, smart transportation, smart environment monitoring and smart tourism.

Wang Tao describes with his contribution “Interdisciplinary Urban GIS for Smart Cities: Advancements and Opportunities” the role of GIS in a Smart City context. He clearly emphasizes the advantages that modern GIS technology brings to the Smart City and underlines this with examples from transportation and mobility, risk management, urban planning, noise mapping and solar energy. This is amended by valuable information about the activities of some government agencies for a geospatially enabled “Smart Singapore”. Under the term “Interdisciplinary Urban GIScience”, he addresses issues from volunteered geographic information collection, cloud computing and SDI and geo-visualization and human–computer interaction. With the help of GIS, cities have already succeeded in transforming their managements to a more efficient level. Yet, much remains to be done to make full use of the potential of modern GIS.

Actual, accurate and complete data are very crucial for designing, modeling and planning in Smart Cities. Geomatics always has played a significant role when it comes to reality-based data acquisition and processing. The present times are characterized by the availability of a great amount of different sensors for a variety of data collection activities. Laser-scanners, both from the aerial and from the terrestrial platforms play a very important role in today’s fast data collection scenery. In his contribution on“Building Reconstruction from Airborne Laser Scanning Data”, Huang Xianfeng gives an account of the state-of-the art in airborne laser point cloud processing for the purpose of generating 3D city models, in particular buildings. The overall process of modelling from point clouds is split up in detection, feature extraction and 3D model generation. A particular point of discussion is the quality control of the results of otherwise automated procedures. Standard methods for quality evaluation are still missing. While raw data acquisition with aerial and terrestrial platforms is very fast today, there are still many bottlenecks in data processing, considering the pipeline from unstructured point clouds to structured surface models.

Another approach for reality-based modelling is via the use of images. In “Increasing Detail of 3D Models Through Combined Photogrammetric and Procedural Modelling”, the authors Stefan Müller-Arisona, Chen Zhong, Xianfeng Huang and Rongjun Qin investigate a new method of combining reality-based 3D models, generated from images, with procedural (generic) modelling in a single workflow, in order to derive high quality/higher resolution hybrid building models. They show how the approach increases the level-of-detail and texture quality compared with using photogrammetry alone, and how their methodology is applicable in practice in terms of a concrete case study based on satellite imagery and terrestrial façade photographs. As project example serve the new-town building towers in Singapore’s Punggol area. In this project context two leading software packages are used: CyberCity Modeller for reality-based modeling and City Engine for procedural modeling.

In recent years, street images have gained enormously on popularity with respect to using them in the model generation process. Images, generated by highly professional Mobile Mapping Systems and/or by the unaware visitor/tourist of a site, are available now for a city in the millions through the Internet. Crowdsourcing is currently in fashion as a replacement of or addition to professional high quality, but slower, data collection. Jan-Michael Frahm, Jared Heinly, Enliang, Zheng Enrique Dunn, Pierre Fite-Georgel and Marc Pollefeys show us in their article “Geo-registered 3D Models from Crowdsourced Image Collections” what can be done in terms of 3D modelling with such huge amounts of partly low-quality images, acquired for non-photogrammetric purposes. They present a detailed algorithmic description of their approach and emphasize that they can process almost three million Internet images of Berlin in less than one day on a single PC. Unfortunately, a quality assessment of the results is missing. For real and meaningful applications, it is mandatory to have a clear understanding of the quality (accuracy and completeness) of the results and the drawbacks of this methodology.

For a variety of applications in a Smart City, knowledge about the underlying terrain is crucial. Some applications, like the modelling of urban floods, require very high accuracy and resolution DTMs. The modelling of terrain has been a topic of research for many years. With the increase in datasets, the issue of computational speed for real-time applications, like in man-made and natural hazards monitoring, becomes crucial. Different forms of terrain models exist. Xie Xiao, Xu Weiping, Zhu Qing, Zhang Yeting and Du Zhiqiang present in “Integration Method of TINs and Grids for Multi-resolution Surface Modelling” a highly efficient hybrid data structure for the seamless integration of multiresolution models. In experiments from highway constructions, using typical real design data sets, the authors are able to achieve accuracy-preserved and real-time availability of results that prove the validity and efficiency of their approach.

We believe, and hope the reader agrees, that we have assembled a very interesting number of Geomatics topics, relevant for various support functions and an efficient management of future Smart Cities.

This first issue of Geo-Spatial Information Science serves as a starting point and “teaser”. Many more and a much wider spectrum of contributions to the topic will hopefully follow.

Here, the co-editor thanks all authors for their willingness to contribute to this journal and for their very cooperative attitude. It was a pleasure for me to work together with this excellent group of experts.

Where does the future take us? It is easy to predict. Those cities that do not change, that do not forge ahead with the use of innovative urban planning, technological and governance models and intelligent use of resources, those that do not follow the concept of smart cities, will be left behind, with all the negative consequences for their population. They will lose financially, miss the best human talents and suffer economically and environmentally. Yet, in forward-looking and future-oriented cities Geomatics will continue to play an important role in this scenery.