Description: This document is the first of three RAILS reports introducing the ongoing work on the pilot cases studies selected to investigate the application of AI methods in different railway domains. This Deliverable specifically focuses on automatic and safe train operations. The following two case studies are being developed: 1) Obstacle Detection and Collision Avoidance, and 2) Cooperative Driving for Virtual Coupling of Autonomous Trains. The discussion includes the description of AI methods, tools, and datasets used in the development of the research activities as well as some hints about expected results and possible criticalities.
Target audience: All rail stakeholders
How it brings us closer to achieving better rail for Europe: The RAILS pilot case studies are conceived as benchmarks for developing proofs-of-concept at the aim of providing insight towards the definition of qualitative and technology roadmaps which could lead to the deployment of AI applications. The ongoing work presented in this document addresses a very innovative research area that is autonomous and cooperative driving in railways. A
comprehensive set of research directions and perspectives are highlighted, that could be explored in addition
to the two case studies described in the deliverable.
More information on this topic: RAILS
Description: The deliverable provides the safety case definition and certification procedure for different technologies introduced in the next generation train control and monitoring system (NG-TCMS). This deliverable covers exclusively the generic parts of the safety case without entering into the specificities of the project-specific architecture. This also means that the hardware components to be selected for safety-critical applications are only specified in terms of their suitability, but no concrete choice is made, for example in the case of the Consist Control Units (CCUs).
Target audience: Operators & Suppliers
How it brings us closer to achieving better rail for Europe: The goal of this deliverable is to prepare a specification to enable the remote driving and command for a number of relevant use cases, carefully selected. Such use cases are:
More information on this topic: TAURO