Machine Learning and Optimization Approaches in Smart Rail Systems
1Beijing Jiaotong University, Beijing, China
2Technical University of Madrid, Madrid, Spain
3University of Liverpool, Liverpool, UK
4South China University of Technology, Guangzhou, China
Machine Learning and Optimization Approaches in Smart Rail Systems
Description
Rail systems are now reaching an important development stage worldwide. High-speed rail transport systems have achieved significant development in Europe and Asia and are beginning to become popular in many countries. Metropolitan transport will have an increasing importance in the coming years to reduce pollution and congestion in cities. Both metropolitan and high-speed railways require the use of advanced signaling and control systems to guarantee and optimize their operation. For these reasons it is necessary to use modern communication and signaling systems for the intelligent control of these railways. In addition, the railway infrastructures use many supplementary systems such as remote control, video surveillance, obstacle detection and operating aids that require the intensive use of information and communication technologies. In all cases the electrical and electronics equipment must have a high quality of service, reliability and availability to fulfill railway requirements.
In the past few decades, machine learning and optimization approaches have become popular methods for solving complicated problems, especially those that have no known algorithms for solving their exact resolution in polynomial time. Unlike the hard computational (classical) methods that focus their entire determination and ability to be precise, and to perfectly model the truth, deep learning and optimization approaches are based on tolerance of inequities, partial and incomplete truths, and lack of certainty. In simple scientific terms, hard methods are driven by nature and how things behave, while deep learning and optimization practices are directed towards humans and the measures taken by their minds to resolve issues. Using "machine learning and optimization" computational methods, one can study, model, and analyze very complex phenomena in smart rail systems.
The aim of this Special Issue is to present a collection of high-quality research papers on recent developments, current research challenges and future directions in the use of machine learning and optimization approaches to realize smart rail systems that are safer, and more efficient. We are soliciting original research and review articles.
Potential topics include but are not limited to the following:
- Machine learning and deep learning in smart rail systems
- Rail system modeling and optimization
- Architectures, optimization algorithms and protocols for data dissemination, processing, and aggregation for smart rail systems
- Rail network information processing, decision making, and optimization
- Railway communications and networking optimization
- Communication system optimization approaches for smart rail systems
- Machine learning and optimization models for train control system in smart rail systems
- Security, privacy, and dependability optimization in smart rail systems
- Rail traffic modeling and optimization, decentralized congestion control and optimization