A convenient and safe transport environment is important for a densely populated city. Five innovative projects led by PolyU researchers have recently received support from the Smart Traffic Fund with a total amount of HK$10.7 million, promising to facilitate the development of intelligent transport that meets Hong Kong’s needs.

The technologies developed in these projects employ AI, deep learning, 3D geo-spatial models, data-driven techniques, etc. to provide reliable and efficient ways to predict various traffic data, which will help enhance the commuting convenience of the public and improve driving safety.

Some of the awarded projects focus on developing predictive management in the areas of shortage of parking spaces, accident-prone black spots and driver’s psychological instability. And there are studies to tackle traffic congestion through the development of estimators or detectors to predict traffic flows and cope with uncertainties.

With a view to achieving “Smart Mobility” in Hong Kong, as one of the areas of Hong Kong’s Smart City Blueprint, PolyU will continue to develop novel technologies and promote transport-related applications to assist the industry and authorities in establishing appropriate strategies.

Below is a summary of the projects awarded the Smart Traffic Fund:

Prediction of Traffic Speed and Volume considering Malfunctioning Detectors using Deep Learning (Prof. Edward Chung, Department of Electrical Engineering)

This project aims to develop a Deep Learning model for predicting near-term traffic speed and volume when some detectors malfunction. The Deep Learning model is also applicable for inputting missing data in offline applications.

Development and Deployment of an AI-enabled Parking Vacancy Prediction Framework using Multi-source Data (Dr Ma Wei, Department of Civil and Environmental Engineering)

This project aims to develop a framework for predicting the short-term parking vacancy for both on-street and off-street parking in Hong Kong. A website and mobile phone-based parking guidance application will then be developed for providing predicted parking vacancy information to the public.

Road Safety Assessment using Advanced Driving Simulation Approach with 3D Geo-spatial Model (Dr Sze Nang Ngai, Department of Civil and Environmental Engineering)

This project aims to develop a 3D geo-spatial model that can be used for safety assessment in driving simulation experiments. An evidence-based decision support tool will be developed for identifying accident-prone locations and recommending safety improvement measures.

Network-wide Traffic Speed-Flow Estimator (Ir Prof. William Lam Hing-keung, Department of Civil and Transportation Engineering)

The project proposes a model-based data-driven approach to develop a network-wide traffic speed-flow estimator for estimating traffic speeds and traffic flows simultaneously.

Investigation of an online data-driven intelligent automation platform for drivers considering the psychological condition instability and behaviours for a sustainable and safe transportation system (Dr Ng Kam-hung, Department of Aeronautical and Aviation Engineering)

The project aims to develop an online data-driven risk-taking behavioural prediction mechanism by identifying the driver’s psychological condition instability using intelligent automation techniques.