The rapid development in machine learning (ML) and deep learning (DL) is largely fueled by Big Data generated by modern social and economic activities, in which transportation plays an integral role. Transformative transportation technologies, ranging from commonplace ride-hailing/ride-sharing and ...
The rapid development in machine learning (ML) and deep learning (DL) is largely fueled by Big Data generated by modern social and economic activities, in which transportation plays an integral role. Transformative transportation technologies, ranging from commonplace ride-hailing/ride-sharing and Micro-mobility services nowadays to Mobility as a Service (MaaS) and connected and automated vehicle technology in the foreseeable future, continue to reshape our socioeconomic landscape and generate tremendous data. This continuously growing Big Data across heterogeneous sources provides a solid data foundation for developing and testing novel ML and DP methods to enable truly smart, effective solutions to the transportation challenges we are facing today. For example, the trajectory data in large scales could be leveraged to gain insight into the behavior of transportation users (e.g., choice on time to travel, destination, and mode) and improve the performance of transportation systems via smart design features and operational strategies. Advancement in computer vision and deep learning (e.g., image and video understanding) has enabled many practical tasks to be automated, such as the collection and analysis of richer transportation data (e.g., pedestrian pose, vehicle classification and re-identification, real-time trajectories, truck taxonomy, logo recognition, etc.) and infrastructure condition data (e.g., various pavement distress types and severities), as well as real-time monitoring of traffic scenes for congestion and incident situations. Congestion could be more effectively managed through smart collaborative driving systems that heavily rely on real-time data sharing among vehicles. Mobility service can be significantly improved by emerging transmodal vehicle systems driven by reinforcement learning algorithms that learn continuously with evolving contexts.
This Research Topic is focused on the new and smart applications of machine learning methods and techniques to improve transportation services and/or address various modern transportation challenges. Technical articles related to the application of machine learning in the following areas are particularly welcome.
● Novel data acquisition, fusion, curation, and analytics in transportation
● Automated vehicle systems and Advanced Driving Assistance Systems (ADAS)
● Smart traffic control algorithms
● IoT applications in event-based management (e.g., traffic incidents, work zone, inclement weather, and special events)
● Road safety analysis with high-resolution data integrated from various sources
● Innovations in image or video-based infrastructure condition assessment (e.g., novel deep learning architectures)
● Cooperative traffic control in connected environments
● Dynamic ridesharing
● Real-time network optimization
● Demand modelling and management using big data
● Operation and management of demand-responsive systems using big data
● Applications of machine learning in behavioral analysis
Data Analytics, Machine Learning, Deep Learning, Computer Vision, Internet of Things, Connected and Automated Vehicles, Intelligent Transportation Systems, Big Data Applications
All contributions to this Research Topic must be within the scope of the section and journal to which they are submitted, as defined in their mission statements. Frontiers reserves the right to guide an out-of-scope manuscript to a more suitable section or journal at any stage of peer review.