News

Press sharing on SCRI smart cities related research

On 26 April, a press conference was conducted at PolyU by Prof. John Shi, Chair Professor of LSGI and Director of Smart Cities Research Institute to share a number of cutting-edge patented technologies and research by his research group. These projects help address various societal issues, including the revitalisation of old buildings, slope safety, prevention and control of the COVID19 pandemic, and the construction of spatial data infrastructure, hoping to provide comprehensive solutions for smart city development in Hong Kong and the Nation.   The project Smart Cities Research Platform received a Gold Medal at this year’s Special Edition 2021 Invention Geneva Evaluation Days- Virtual Event.    Other research projects shared:  Three-dimensional (3D) Mobile Mapping System: Providing accurate 3D maps to support wide smart city applications AI-based Landslide Recognition: Reporting landslide and facilitating disaster control Spatiotemporal Prediction of COVID-19 Onset Risk: To help public health agencies formulate more precise prevention and control strategies

SCRI Scholars Publish the Book Urban Informatics

The new book entitled Urban Informatics has been edited by Prof. Wenzhong Shi (Chair Professor of LSGI), Prof. Michael Goodchild (University of California, Santa Barbara), Prof. Michael Batty (University College London), Prof. Mei-Po Kwan (The Chinese University of Hong Kong) and Dr. Anshu Zhang (Research Assistant Professor of LSGI), was published on 8 April 2021. This is open access and a milestone book which systematically introduces the principles of emerging urban informatics and its wide applications in enabling cities to function more efficiently and equitably, to become ‘smart’ and ‘sustainable’.   The milestone book brings together over 140 authors from more than 40 world-leading research teams. The book is organized into five main parts: urban science, urban systems and applications, urban sensing, urban big data infrastructure, and urban computing. For each field which are covered under the five parts, the book offers a comprehensive review and a detailed technical introduction to the principles, methods and tools that form the core of urban informatics. It also outlines ways in which these methods and tools can be used to inform the design, policy, and management of urban services as well as ways in which cities can be planned to become more efficient with a greater concern for environment and equity.    The open access book can serve as a reference book for the researchers and professionals, as well as a textbook for relevant postgraduate and undergraduate courses.    Link for download: https://www.springer.com/gp/book/9789811589829 

SCRI scholars ranked as the world’s top 2% most-cited scientists by Stanford University

We are glad to share that four academic staff from SCRI are listed among the top 2 % scientists in the world according to a report by Stanford University in early 2021. The listed SCRI scholars are: Prof. John Shi, Chair Professor of LSGI and Director of SCRI, Dr. Xiaolin Zhu, Assistant Professor of LSGI and Member of SCRI   The report was prepared by a research team led by Prof. John Ioannidis of Stanford University. According to the team, the publicly available database of scientists among the world that provides standardized data on citations, h-index, co-authorship adjusted hm-index, citations to papers in different authorship positions, and a composite indicator. The SCRI scholars have demonstrated substantial influence in various disciplines through the publication of multiple highly cited papers ranked according to citation impact in 2019. While Prof. Shi is also listed in the top 2% most-cited scientists for career-long impact from 1996 to 2019. Disciplinary differences in terms of citation norms are taken care of since the list identifies the top 2% of scientists in their own areas of specialty.  Congratulations to all of the listed scholars! 

SCRI Research: Logistics Big Data for Intra-Urban Goods Movement Patterns

This research proposes an analytical framework for exploring intra-urban goods movement patterns by integrating spatial analysis, network analysis and spatial interaction analysis. Using daily urban logistics big data (over 10 million orders) provided by the largest online logistics company in Hong Kong (GoGoVan) from 2014 to 2016, we analyzed two spatial characteristics (displacement and direction) of urban goods movement. The origin–destination flows of goods were used to build a spatially embedded network, revealing that Hong Kong became increasingly connected through intra-urban freight movement. Finally, spatial interaction characteristics were revealed using a fitting gravity model. Workflow of The Empirical Analysis   1. Spatial Characteristics of Goods Movement We observe that the probability distributions for datasets of different years displayed similar trends and could be well fitted by a bimodal Weibull distribution. Specifically, two maximum points exist, namely, 6 km and 22 km. This means that the count of goods movement did not decrease monotonously with distance between origin and destination. Interesting to note is that the intra-urban freight movement displacement distributions failed to follow an exponential or power law distribution. According to the depiction in subsection on the measurement of spatial characteristics, we can conclude that average freight distance gradually increased. We can also conjecture that increasingly improved transportation networks provided greater convenience for freight, especially long-distance freight. Displacement distributions in different years: (a) probability distributions for the datasets; (b) cumulative probability distributions for the datasets. Direction distributions over 3 years   2. Characteristics of the spatially embedded network of goods movement We constructed a spatially embedded network based on origin–destination flow between units and conducted network analysis. By comparing the network properties for the networks in the years 2014, 2015 and 2016, we observed that Hong Kong became increasingly connected from the perspective of logistics. Furthermore, we investigated the distributions of degree and strength of nodes and examined the correlation between them. We found that their relationships could be well fitted by exponential functions, and all values of goodness-of-fit R2 reached 0.72 or higher. In other words, the freight flows between subdistricts could be estimated by the connectivity of subdistricts at the aggregate level. Freight movement flows for network construction Cumulative probability distributions and correlations of degree and strength   3. Characteristics of Spatial interaction We explored the spatial interaction characteristics of intra-urban freight movement and how the interaction flows were related both to the population (or total trips) of the origin and destination and to the distance between subdistricts by fitting the gravity model. The significant linear relationships between interaction flow and the product of the subdistrict populationsPiPj were observed by fitting the general form of the gravity model. In addition, we found that gravity was suitable for predicting goods movement flows by comparing the estimated interaction flows and observed interaction flows. However, the distance decay parameterβ was significantly smaller than that of human mobility patterns. We concluded that the spatial interaction of goods movement was not substantially influenced by the distance between origin and destination.   Comparison of estimated interaction flows with the observed interaction flow based on the datasets for the years 2014, 2015 and 2016   REFERENCE Zhao, P., Liu, X., Shi, W., Jia, T., Li, W., & Chen, M. (2020). An empirical study on the intra-urban goods movement patterns using logistics big data. International Journal of Geographical Information Science, 34(6), 1089-1116. Contact: Dr. Xintao Liu (xintao.liu@polyu.edu.hk)  

SCRI scholars excel at Geneva Inventions Expo

In a special virtual edition of the 48th International Exhibition of Inventions of Geneva, the Inventions Geneva Evaluation Days were held from 10 to 14 March 2021. Gold Medals have been awarded to Prof. John Wenzhong Shi and Dr Charles Man-sing Wong of the Department of Land Surveying and Geo-Informatics (LSGI) for their smart city platform and smart monitoring system for urban tree management respectively. Project “Seamless Navigation in Urban Environment” led by Prof. Wu Chen, Head of LSGI, and funded by Logistics and Supply Chain MultiTech R&D Centre (LSCM), received the Silver Medal. They were among the seven PolyU awardees and projects honoured at the virtual event.    Smart City Platform: A Comprehensive System for Spatial Data Infrastructure PI: Prof. John Wenzhong Shi (LSGI) The smart city platform can be used to create digital city replicas for acquiring insights into urban situations, testing solutions and conducting technological research. Incorporating 3D city modelling, AI-based urban object cognition, web-based visualisation and analytics technologies, it enables seamless fusion of massive geometrical information, 3D LiDAR data, image data and spatial big data from various sources, including public and private agencies, to provide a realistic and accurate representation of a city. A 3D spatial data acquisition system has been specially developed to enhance both outdoor and indoor 3D environment of buildings. It can 2 identify indoor objects and automatically reconstruct a digital model (Building Information Model) of the indoor environment from raw LiDAR data.   Smart Monitoring System for Urban Tree Management PI: Dr Charles Man-sing Wong (LSGI) The smart monitoring system for urban tree management makes use of smart sensing technology and geographic information systems to monitor the stability of local trees. Tailor-made sensors are installed on the lower trunks of selected urban trees to monitor their tilting angles. The data collected has facilitated timely mitigation measures for sustaining longer tree lives. Our researchers devote their innovative research to changing human life. This recognition gave strong support for them to advance their research in related area and application.    Seamless Navigation in Urban Environment PI: Prof. Wu Chen (LSGI) This research developed technology to improve mobile phone positioning accuracy with reasonable cost. It enables the positioning to achieve an accuracy of 2 metres in open areas, and less than 10 metres in dense urban areas, for mobile phone-based and location-based services. It uses a positiondomain DGNSS (Differential Global Navigation Satellite System) platform with an accuracy of 2 metres, the multipath mitigation engine with the integration of GNSS observation, Microelectromechanical Systems (MEMS) inertial sensors and a 3D map for greater accuracy in positioning. 

SCRI Research: Smart City Platform 4D comprehensive spatiotemporal data infrastructure

In a smart city, intelligent machines communicate with each other via the Internet of Things (IoT) and their sensors collect data every second.  To facilitate the smooth running of a smart city, such IoT data are geotagged, so that we know where the data are captured. But in a vertical city like Hong Kong, geotagging alone is not enough as there could be tens of floors stacked up at the same spot on a map.  A 3D model, or better still, a 4D model that incorporate 3D geo-information from various sources plus temporal dynamic models would cover more grounds for various smart city applications.   In light of this, Prof. John Shi, Head of the Department of Land Surveying and Geo-Informatics, led a research team to develop the Smart City Platform, a 4D comprehensive spatiotemporal data infrastructure that consolidates spatiotemporal data from various sources under one roof, with a web-based visualisation interface for easy reference.  The team also devised a portable 3D LiDAR system to acquire spatial data of older buildings and established an international standard for underground utility model. Compatible with various data sources A city is a complex cluster of buildings, roads and open spaces, where various activities take place at the same time. Mapping out all these infrastructures and activities are essential to make a city smart.  Although experts have invented systems to collect various data, such as 3D city modelling, building information modelling (BIM) and IoT sensors, these systems run on different platforms collecting data in different formats.  Users have to use different software to retrieve different spatial data, making cross-referencing tedious and difficult. Prof. Shi thus developed the Smart City Platform.   “Our system can read data in various formats, including 3ds, obj, skp, shp and gdb files.  It can also process data from IoT sensors directly without the need of conversion.  All data are consolidated on one platform, and the results are visualised in 3D format accessible to any web browser.  In other words, you don’t need to install any software to read the 3D map and the layers of information embedded therein,” Prof. Shi explained.  The system incorporates indoor, outdoor, above-ground and underground 3D spatial information, plus spatiotemporal dynamic models. 3D LiDAR spatial data Besides consolidating data from various sources, the system developed by Prof. Shi’s team also captures its own raw data with 3D LiDAR. “There are many sources that provide the location and exterior configuration of a building, such as satellite positioning and 3D city modelling.  But when it comes to its interior structure, we have to access its BIM which is rarely available in older buildings.  So, we developed a portable 3D LiDAR spatial data acquisition system to collect their interior spatial data,” Prof. Shi said.  The system is made up of a backpack with a metal frame where laser emitters and sensors are attached.  The operator puts the system on his back and walks around the interior of a building to capture the position and measurement of every dot and line with simultaneous localisation and mapping (SLAM) technology.  The post-processing software then combines the data collected to re-create the interior 3D space automatically.  That means BIM can be built for older buildings for smart city functions to operate seamlessly. Establishing underground utility model standard A smart city not only runs above ground, but also underground – there is a vast network of pipes and cables that carry water, gas and electricity to each building, while removing waste.  Before digging underground, one must accurately locate all utilities. Prof. Shi said, “Each utility company keeps good records of its own pipes or cables underground.  But it may not have access to the records of other utility suppliers.  There wasn’t even an international standard on how to document such underground utility network in the past.  As our data infrastructure also covers underground spatial data, we figured we need to establish such a standard and we have submitted the specifications to the Open Geospatial Consortium (OGC) for approval.” Applications The Smart City Platform is versatile and has great potential in various applications.  It is suitable for urban planning, virtual test-bedding and visualisation of proposed developments.  By analysing future urban development on the platform, developers can predict its impact on traffic and environment.  Businesses also benefit from its functions such as spatiotemporal mapping of social media engagement and supply chain logistics management.  Amid COVID-19 pandemic, the system was able to predict the level of risk among various DSE examination centres based on distribution data of confirmed cases so that candidates might take extra precautions if necessary.  The possibilities are endless.  Source: https://www.polyu.edu.hk/ife/corp/en/publications/tech_front/22053 

The 2nd International Conference on Urban Informatics, ICUI 2019