City-scale high-resolution traffic datasets with refined networks for hierarchical traffic control

Area of Science:

• Transportation Science
• Urban Planning
• Data Science

Background:

• Urban traffic control is hindered by a lack of large-scale, high-resolution datasets for diverse network topologies.
• Existing datasets often lack the granularity needed for advanced traffic management strategies.

Purpose of the Study:

• To present comprehensive traffic datasets for benchmarking urban traffic control strategies.
• To provide high-fidelity data supporting the development of hierarchical traffic control architectures.
• To enable realistic traffic simulations across various city scales and network configurations.

Main Methods:

• Collected and processed traffic data from five cities with varying scales and intersection layouts.
• Developed a refined network representation for the Xuancheng dataset, including complete vehicle path recording.
• Incorporated vehicle type information and long-term historical trip data with demand pattern analysis.

Main Results:

• Generated multi-level traffic state data (vehicular, lane-intersection, network levels).
• The Xuancheng dataset features hundreds of intersections with high spatiotemporal resolution.
• Datasets capture periodic and holiday-induced demand variations for realistic simulation.

Conclusions:

• The presented datasets address the critical need for high-quality data in urban traffic control research.
• These resources facilitate the development and evaluation of advanced, multi-level traffic control strategies.
• The datasets support the creation of realistic simulation environments for testing traffic management solutions.