With the rapid developments of sensors, artificial intelligence, and data mining techniques, intelligent transportation systems (ITS) have emerged as a revolutionary paradigm to improve traffic safety and efficacy. Different types of sensors with an increasingly attractive price-to-performance ratio have encouraged wide and successful applications in ITS.

However, due to the rapidly increasing amount of sensed traffic data, it inevitably becomes difficult to guarantee high resilience and efficiency in ITS-applied scenarios. Therefore, the ongoing transportation revolution (especially autonomous transport systems) puts forward high requirements for the development of advanced traffic modeling and optimization techniques. Benefiting from the emerging intelligent techniques, it is able to make data-driven traffic modeling and optimization more reliable and practical in complex transportation conditions. The safety, security, sustainability, and efficacy of multi-modal traffic could be promoted accordingly. Although tremendous progress has been achieved in traffic modeling and optimization methods in ITS, both academia and industry are still facing several challenging problems which hinder further advances in ITS development: how to improve the sensor data quality for different types of sensors (e.g., radar/vision/positioning sensors, etc.) under complex traffic conditions; how to robustly perform multi-source data fusion for identifying the most influential factors affecting traffic safety and sustainability; how to properly characterize the traffic behavior and accurately estimate the traffic situational awareness; how to robustly and accurately solve the large-scale optimization problems arising in traffic organization and transportation planning; and finally, how to guarantee the effectiveness and efficiency of autonomous transport devices using data-driven computational methods.

This Special Issue will focus on recent data-driven traffic modeling and optimization methods in ITS, which address the original theoretical developments and practical applications. We especially welcome high-quality original research and review articles which cover a broad range of topics related to data-driven methods and their potential applications in traffic modeling and optimization. Research on the recent progress of traffic modeling and optimization in road, rail, water and air transport, etc., will also be considered.

Potential topics include but are not limited to the following:

• Traffic video/image quality enhancement under adverse weather conditions
• Missing traffic data imputation
• Multi-sensor perceptual data acquisition, fusion, and analysis
• Visualization and visual analysis of traffic data
• Intelligent methods for large-scale optimization problems
• Supervised/unsupervised/self-supervised learning in traffic modeling and optimization
• Data-driven prediction of traffic flow and movement trajectories
• Data-driven intelligent collision avoidance
• Data-driven moving object detection, recognition, and tracking
• Data-driven traffic behavioral modeling and anomaly detection
• Traffic situational awareness and safety management
• Positioning and navigation for autonomous transport devices
• Motion planning for autonomous transport devices