The Hong Kong Polytechnic University (PolyU) is committed to creating a safer, smarter future through impactful innovation. In the 26th batch of the Smart Traffic Fund, two PolyU projects have received total funding of approximately HK$9.87 million. These projects focus on real-time bus dispatching systems and vision-language model-based digital mapping technologies respectively, and aim to enhance public transport efficiency and road network management.
Led by Dr. Wenbo FAN, Research Assistant Professor of the Department of Electrical and Electronic Engineering of PolyU, the project “A Real-Time Bus Dispatching System for Improving Bus Operational Efficiency through Arrival Time and Headway Regulation” secured funding of approximately HK$2.27 million for 24 months.
This project will develop a bus dispatching system to enhance bus operational efficiency through the real-time regulation of bus arrival times and headways. By analysing large-scale bus operation data, the system will apply advanced optimisation algorithms to provide bus captains with real-time speed recommendations. By jointly optimizing bus headways and driving efficiency, the project aims to maintain regular bus arrival time and headways, and reduce bus bunching, passenger waiting time, travel time variability, and unnecessary acceleration and braking, while improving bus fleet utilization.
Led by Prof. Wei MA, Associate Professor of the Department of Civil and Environmental Engineering of PolyU, the project “Development of Vision-Language Model-based Lane-Level Digital Map Modelling Technology for Establishing Lane-Level Road Network Digital Maps” secured funding of approximately HK$7.6million for 24 months.
This project aims to apply vision-language models and artificial intelligence (AI) to develop modelling technologies for generating lane-level road network digital maps. By integrating multi-source geospatial and road data—including street-view imagery, aerial/satellite imagery, government road-network datasets and open-source maps—the platform generates a lane-level digital road network map for Hong Kong.
By coordinating specialised AI agents for data interpretation, lane-feature extraction and topology modelling, the project creates accurate, machine-readable lane-level digital maps that provide precise lane-level characteristics, traffic flow direction information and connectivity, etc.
PolyU has long been committed to vehicle-related research and technology application, with 35 projects supported by the Smart Traffic Fund to date. This achievement underscores the University’s contribution to advancing transportation technology innovation.
The Smart Traffic Fund provides funding support to local organisations and enterprises for conducting research and applying innovation and technology with the objectives of enhancing commuting convenience, enhancing efficiency of the road network or road space, and improving driving safety.
