Matthew BoyleHong Kong Polytechnic University (PolyU) researchers are using artificial intelligence (AI) and geospatial analysis to tackle industrial growth and climate change challenges. Their GeoAI approach combines AI and geospatial technologies to provide insights into natural and social environments.
GeoAI has transformed building monitoring, disaster detection, and energy consumption estimation by leveraging thousands of measurable variables. PolyU researchers have developed models like the impervious surface area (ISA)-based urban cellular automata (CA) model to simulate urban growth and promote green development.
Their work, published in journals like Cities, offers insights into urban sprawl patterns, aiding in global urban planning efforts. GeoAI also enhances real-time traffic management and ride-hailing services through machine learning and deep reinforcement learning techniques.
Led by Professor Qihao Weng, the PolyU Research Centre for Artificial Intelligence in Geomatics (RCAIG) aligns with UN Sustainable Development Goal 11 to create sustainable cities. The RCAIG focuses on advancing AI methodologies for geomatics, using satellite observations and real-time data for informed decision-making.
Harnessing GeoAI for Urban Resilience and Sustainable Development: Insights from Hong Kong Polytechnic University Researchers
To address emerging issues relating to industrial growth and climate change, researchers at Hong Kong Polytechnic University are using AI and informatics.
The researchers want to develop geospatial and artificial intelligence ( GNI ) technologies to provide solutions and insights into dynamic changes occurring in both natural and social environments. The applications of GeoAI are rapidly expanding across several fields, encompassing transportation, cosmopolitan and public safety, planning, climate change and natural disasters.
By utilizing thousands of measurable variables, GeoAI has revolutionized building monitoring. This ability, which shows how well it can automatically recognize public patterns in buildings, such as color and shape, is an illustration of this. This technology is critically applied to detect disaster- destroyed buildings, retrieve building height, identify structural changes, and estimate building energy consumption. In consequence, GeoAI has come to be a common tool for more effective and incisive building monitoring.
These technologies aim to improve the resilience of urban areas to withstand and recover from a variety of challenges, such as severe heatwaves, while supporting the well-being and sustainable development of the urban population.
Advancing Urban Planning with GeoAI: RCAIG’s Impervious Surface Area-based Cellular Automata Model for Sustainable Development
A research team from the RCAIG has created an impervious surface area ( ISA)-based urban cellular automata ( CA ) model that uses annual urban extent time series data from satellite observations to simulate the fractional change of urban areas within each grid. The model provides more detailed insights than conventional binary CA models by describing the traditional patterns of urban area growth under various levels of urbanization. This demonstrates how much of its potential is for promoting green development.
The study, which was conducted by Ms. Wanru He, an RCAIG a postgraduate research assistant, and the team, was published on Cities. Their model properly captures urban sprawl’s dynamics with significantly improved computing efficiency and performance, and it makes it possible to model urban growth at regional and global levels, under various future urbanization scenarios.
GeoAI uses machine learning and deep learning to analyze complex data, providing applications like real-time traffic management. Through the integration of diverse data modalities, such as text, images, and knowledge graphs, GeoAI enables appropriate traffic flow prediction, route optimisation, accident warnings, and the planning of an effective traffic network. Therefore, this contributes to the advancement of bright traffic management.
Revolutionizing Ride-Hailing Services: RCAIG’s Multi-Agent Approach for Efficient Order Matching and Vehicle Repositioning
The research team at the RCAIG has developed a multi-agent order matching and vehicle repositioning ( MAMR ) approach to increase the effectiveness of ride-hailing platforms and achieve intelligent management of their services. This cutting-edge technology aims to coordinate the supply and demand for ride-hailing services, inevitably aiming to increase their overall efficiency.
This approach offers a revolutionary approach to address two essential elements of effective ride-hailing services. Second, it addresses order matching by effectively distributing orders to accessible vehicles. Third, it incorporates strategic vehicle repositioning, strategically deploying empty vehicles to regions with possibly high demand. This innovation, which is based on multi-agent deep reinforcement learning ( MARL), addresses the challenging planning issues in transportation and provides a news perspective on the long-term spatiotemporal planning issue. The study, titled” Multi-agent reinforcement learning to unify order-matching and vehicle-repositioning in ride-hailing services,” was published in the International Journal of Geographical Information Science led by Ms. Mingyue XU, a second RCAIG researcher and the team. The study’s outcomes, such as decreased driver empty time and passenger rejection rates, outperformed the study. The RCAIG and the POLEIS at PolyU are dedicated to conducting research in a variety of fields, including urban building and energy, urban safety and security, environmental monitoring and conservation, and urban resilience and public health, with a focus on geospatial artificial intelligence ( GeoAI ). This aligns with the 11th United Nations Sustainable Development Goal ( SDG11 ), which aims to create inclusive, safe, resilient, and sustainable cities and human settlements.
Advancing Geomatics with AI: PolyU’s RCAIG Spearheads Research for Urban Resilience and Public Health
The PolyU Research Centre for Artificial Intelligence in Geomatics (RCAIG ) was established in order to promote the development of AI methodologies and technologies for geomatics and their applications in urban areas. Professor Qihao Weng, Chair of the Department of Land Surveying and Geo-Information, and Global STEM Professor, is also the Chair of Geomatics and Global Professor.
Professor Weng said,” Our research encompasses a broad spectrum of subjects in the fields of earth observations and geoinformatics. Satellite observations are valuable resources for our community because they depend on satellite imagery, videos, and data for making informed decisions about industrial resilience and public health. Satellite observations, for instance, aid in the development of industrial flood monitoring algorithms and help us understand the effects of extreme heat waves on population exposure. Actual- time data acquisition even facilitate applications in traffic conditions, air quality, nature disasters, population movement and urban land use.
