Artificial Intelligence in Railway Transport: Taxonomy, Regulations, and Applications

Abstract

Artificial Intelligence (AI) is becoming pervasive in most engineering domains, and railway transport is no exception. However, due to the plethora of different new terms and meanings associated with them, there is a risk that railway practitioners, as several other categories, will get lost in those ambiguities and fuzzy boundaries, and hence fail to catch the real opportunities and potential of machine learning, artificial vision, and big data analytics, just to name a few of the most promising approaches connected to AI. The scope of this paper is to introduce the basic concepts and possible applications of AI to railway academics and practitioners. To that aim, this paper presents a structured taxonomy to guide researchers and practitioners to understand AI techniques, research fields, disciplines, and applications, both in general terms and in close connection with railway applications such as autonomous driving, maintenance, and traffic management. The important aspects of ethics and explainability of AI in railways are also introduced. The connection between AI concepts and railway subdomains has been supported by relevant research addressing existing and planned applications in order to provide some pointers to promising directions.

Automatic Generation of Domain-Aware Control Plane Logic for Software Defined Railway Communication Networks

“…In the context of next generation communication networks applied to intelligent transportation systems, we discuss in this paper the advantages in applying the SDN paradigm to future railways. In fact, next generation smart-railways will be increasingly based on novel communication and networking paradigms, including the Internet of Things [14] requiring high performance and bandwidth due to the need for manipulating large amounts of data to support machine learning applications [4], high reliability and low latency to enable emerging signalling paradigms such as moving block and virtual coupling [8], as well as flexibility, dynamic context-awareness, adaptation, scalability, and support for advanced multimedia services.…”

Safe and Secure Future AI-Driven Railway Technologies: Challenges for Formal Methods in Railway

“…We envision to improve this situation by developing verification techniques that provide explainability or guarantees for AI-based systems in the specific safety-critical domain of railway systems, because, as Bešinović et al put it, "although AI is still in its very infancy for the railway sector, there is certain evidence showing that its potential should not be underestimated" [24]. To this aim, we will first need to identify the state of the art of formal methods techniques developed for and applied to systems with AI-based components in the specific setting of transport systems (railways, but also automotive [97,99]) and of safety certification.…”

“…Our focus on AI integration aligns well with the planned successor of Shift2Rail, Europe's Rail Joint Undertaking (EU-Rail), which will specifically focus on digitalisation and automation. So far, there are very few research projects and white papers (cf., e.g., [24,2]) that look into the integration of AI into the Railway domain. Notable exceptions are the RAILS project (Roadmaps for A.I.…”

Intelligent Energy-Efficient Train Trajectory Optimization Approach Based on Supervised Reinforcement Learning for Urban Rail Transits

“…With the enabling of AI [24], reinforcement learning (RL) has been a powerful methodology for addressing decision-making problems in real-time TTO with disturbances. Yin et al [25] developed two intelligent train operation algorithms, and the simulations showed that the RL-based algorithm is capable of adjusting trip time dynamically between two stations.…”