|
Dear reader,
Artificial intelligence is increasingly finding its way into the operational control of rail systems – and the first concrete results are already visible. Deutsche Bahn's AI-assisted dispatching system, deployed across the Stuttgart and Rhine-Main S-Bahn networks since 2021 and in Munich since 2022, processes around 500 pieces of operational data per minute in real time and has prevented more than 58,000 minutes of delays - a step in the right direction, but with Deutsche Bahn’s delays totalling 217 million minutes in 2024, it’s a drop in the ocean. In maintenance, the startup KONUX monitors over 1,300 switches in collaboration with Deutsche Bahn using IoT sensors and AI-based analysis – reducing failures and costs by up to 25 per cent.
These examples illustrate what is at stake: AI in rail operations is no longer a vision. It is becoming an operational reality, and with it come both significant opportunities and serious questions that the industry must address. From predictive maintenance and automated timetable management to AI-assisted traffic control, the potential applications are broad and momentum is growing.
At the same time, the integration of AI into safety-critical infrastructure raises fundamental questions. How do we ensure transparency and explainability in automated decision-making? Who bears responsibility when an AI-driven system makes an error? And how do rail operators manage the transition in terms of workforce skills, regulatory compliance and cyber resilience?
These are not merely technical questions – they are strategic ones. Companies that engage with them early will be better positioned to shape the transformation rather than react to it. Those that wait risk being caught off guard by regulatory requirements, competitive pressure or operational gaps.
Against this backdrop, reliable market intelligence and continuous monitoring of technological developments are more important than ever. The ability to track emerging AI applications, assess their maturity and understand their implications for your own market position is becoming a core strategic capability.
Best regards,
Your RMR-Team |
|
|
|
| |
The RemODtrAIn consortium, led by Siemens Mobility, is developing and testing a secure remote control system for train operations in depots, combined with a modular, AI-based obstacle detection module. The project builds on findings from previous initiatives such as AutomatedTrain and safe.trAIn and consolidates the positive collaboration with Deutsche Bahn. As part of the project, an ICE 4 will be equipped with the latest 5G technology, allowing the train to be controlled remotely from a central control centre on the depot premises.
The focus is on train availability, depot movements and stabling operations. The vehicle sensors are designed for universal use in all operating modes. The consortium thereby also addresses the challenge of the shortage of train drivers and aims to further develop automated and remote-controlled train operation. Obstacle detection is being evaluated under daily operating conditions on the S-Bahn Berlin network, while communication performance is being tested at the Smart Rail Connectivity Campus in Annaberg-Buchholz using a regional Desiro Classic.
Twelve organisations are contributing to the project, including Siemens Mobility, Siemens AG, DB AG, DB Fernverkehr, DB Systemtechnik, DB RegioNetz Infrastruktur, MIRA GmbH, the Smart Rail Connectivity Campus, the German Aerospace Center (DLR), and the Technical Universities of Berlin, Chemnitz and Munich. DLR The project is funded with €17 million by the EU and Germany's Federal Ministry for Economic Affairs and Energy. Vehicle validation is planned for 2028. |
|
| |
|
Track monitoring is one of the most challenging functions in driverless operation on open railway systems. Future safe computer vision systems are intended to solve this task. However, no generally accepted standardised procedure currently exists for demonstrating the safety of automated track monitoring. Future systems will inevitably incorporate elements of machine learning and artificial intelligence, yet the software standard EN 50657 cannot be applied to them. Since AI system components are less transparent than conventional software from a verification perspective, ways must be found to make their functioning sufficiently comprehensible to allow their safety and reliability to be verified.
The aim of the XRAISE project is to identify and evaluate concrete approaches by which methods of Explainable AI (XAI) can support the safety demonstration for driverless operation in open railway systems – for example by resolving currently open steps in the safety case or by reducing the amount of data required for it. The project results are intended to contribute to the development and standardisation of a practical procedure for future safety demonstrations in driverless rail operation.
The project runs for 24 months from February 2024 and is being carried out by a consortium of PECS-WORK GmbH, EYYES Deutschland GmbH, and TU Dresden. | |
|
|
|
|
Research Results Published | |
|
| |
The German Centre for Rail Transport Research (DZSF) has published the results of a study examining the potential of currently available laser scanning systems – both unmanned (ULS) and terrestrial (TLS) – for use in the rail track environment. The research covered three application areas: monitoring of gravitational mass movements, vegetation condition monitoring, and monitoring of structures and infrastructure elements.
Test measurements were conducted at three sites, including a steep slope along the tracks in Orxhausen (gravitational mass movements), a vegetation survey in Rittierode, and structural assessments at the Luhetal Viaduct in Greene and the Gandetalbrücke in Bad Gandersheim.
Key findings include: TLS and ULS each offer distinct advantages depending on the application. For structure and infrastructure element monitoring, TLS proved more suitable due to its higher point density and accuracy. For large-scale tasks such as terrain capture and vegetation monitoring, ULS is the more efficient choice due to its greater range and shorter data capture time. Both systems were found capable of supporting rail infrastructure monitoring – terrain modelling, vegetation analysis and deformation detection – and can be used in combination where appropriate.
The study also found that automated classification and object recognition in combination with laser scanning is a promising direction, with classification algorithms becoming increasingly accurate. Laser scanning cannot yet replace on-site inspections by specialists, but can serve as a valuable complementary tool for decision-making and maintenance planning. |
|
|
| |
|
The German Centre for Rail Transport Research (DZSF) has published the final report of a study examining how digital diagnostic applications can be methodically integrated into existing regulatory frameworks to optimise and partially automate maintenance processes.
The study analysed several diagnostic applications using Failure Mode and Effects Analysis (FMEA), including Continuous Track Monitoring (CTM), the DIANA switch drive diagnostics system (WDS), and video monitoring systems. The results show that CTM and DIANA in particular can provide valuable additional information on infrastructure condition, enabling better-informed and more efficiently planned maintenance decisions. At the same time, none of the technologies examined is currently capable of fully replacing existing inspection procedures. Their greatest benefit lies in supporting existing processes and improving the quality of decision-making.
The study concludes that the integration of Condition Based Maintenance (CBM) and Predictive Maintenance (PM) into rail maintenance regulatory frameworks is fundamentally feasible and offers significant potential for more efficient, data-driven processes. However, further methodological development and structural adaptation are required. Key recommendations include a systematic review of existing regulations, comprehensive cost-benefit assessments of the technologies, improvements in data quality and utilisation, and the targeted further development of new diagnostic systems. |
|
|
| |
The FP4-Rail4EARTH project aims to improve the sustainability performance of Europe's railway system and contributes to the objective of a climate-neutral Europe by 2050. As part of Europe's Rail Flagship Area 4, the project focuses on advancing greener technologies across rolling stock, infrastructure, and their related sub-systems.
Three new deliverables were published in March 2026, covering three distinct research strands. In the field of aerodynamics, researchers are combining advanced computational simulations with wind tunnel testing to analyse airflow behaviour around regional trains, with a particular focus on roof components and pantographs. By improving the aerodynamic design of trains and key components such as roof equipment and pantographs, this work contributes to reducing energy consumption, operational costs, and CO₂ emissions in railway operations.
On energy storage, the project evaluates several technologies for deployment in railway applications, including lithium-ion batteries, supercapacitors, flywheel energy storage systems, and Superconducting Magnetic Energy Storage (SMES), and proposes an innovative method to integrate energy storage systems into existing static frequency converters, enabling peak power shaving and improved utilisation of regenerative braking energy.
The third strand addresses alternative propulsion systems. Pre-standardisation activities support the deployment of battery and hydrogen trains, including the identification of standardised battery interfaces and the development of energy management strategies for alternative drive trains. |
|
|
| |
The FP2-R2DATO project aims to improve rail safety and efficiency by developing Automatic and Autonomous Train Operation (ATO) technologies that are moving toward higher Technology Readiness Levels.
Deliverable D6.5 synthesises results from previous projects, like the Shift2Rail X2Rail4 project, and integrates new findings generated by the FP2-R2DATO project. Its objective is to provide a comprehensive description of advanced GoA3/4 ATO functionality, detailing core functions of automated and autonomous train operation, interfaces and interactions with other key subsystems like Automatic Train Protection (ATP), Train Control and Monitoring System (TCMS) and trackside signalling equipment. Additionally, a semiformal modelling method is used enabling railways operators and suppliers to work together using a shared tool and system architecture.
By improving and harmonising the ATO GoA3/4 specifications, this work supports the development of future automated and autonomous train operations, contributing to increased capacity, improved punctuality and reduced operational costs while leading to a rail system that is more interoperable, energy efficient, digital and automated. |
|
|
| |
The Hyper4Rail project aims to harmonise and realise a concept design of the hyperloop system (TRL 2) and to validate the subsystem technologies required for the development of transport systems in a low-pressure environment (TRL 4), defining a common roadmap for the integration of hyperloop technology into the Trans-European network.
Deliverable D2.1 identifies critical transportation challenges in Europe and identifies opportunities for hyperloop as a new potential transport solution. The document serves as a first step in the Hyper4Rail project in developing the harmonised hyperloop system; several proposed hyperloop solutions and underlying technologies have been identified.
This new transport solution is expected to contribute to strengthening the European transport system by reducing congestion and providing a potentially fast and environmentally friendly transport mode in a low-pressure tube. |
|
|
|
|
Planned Research Projects | |
|
|
|
Perception systems for automated driving in GoA 3 and GoA 4 are currently the subject of intensive R&D work and are generally based on a combination of Lidar and radar sensors as well as RGB and infrared cameras. For testing and development – for example when training AI models – sensor data with a suitable multi-sensor configuration is required. In addition to this data, annotations are of crucial importance as ground truth for both testing and training purposes. Annotated sensor data from dangerous or critical scenarios is also needed, but is not yet freely available.
In this project, sensor data representing critical and hazardous scenarios from the perspective of a perception system in automated rail traffic will be recorded and annotated. This data will serve to develop and test automated systems and their subcomponents, expanding the existing open sensor data base (e.g. OSDAR23) to include the aspect of critical scenarios. The result will be an annotated multi-sensor dataset of dangerous or critical scenarios for ATO development, available for use in further research including by the DZSF itself.
Project duration: 30 months | Planned project start: Q1 2026
|
|
| | | |
The railway sector is rapidly advancing in digitalisation, driven by the growing need to integrate modern IT technologies already established in commercial environments (Commercial Off-The-Shelf; COTS) into safety-critical rail applications. Key areas include innovative communication infrastructures such as Vehicle-to-Vehicle communication and the use of IP-based network protocols in digital interlocking technology. However, significant challenges remain – both in unresolved technical questions and in the constraints posed by existing standards and approval processes.
The aim of the project is to analyse and assess core cybersecurity requirements for the digital railway: determining the usability and security of these technologies under the specific conditions of railway operations, identifying the requirements for their approval, and uncovering existing development potential. Potential risks of a wider rollout can thereby be identified and mitigated at an early stage. The results are intended to feed into the further development of norms and standards and to serve as a scientifically sound basis for approval authorities and companies alike.
Project duration: 36 months | Planned project start: Q1 2026 | |
|
|
|
 |
| Strategic growth
Increase your market presence and generate qualified leads through in-depth market analysis, efficient process and innovation management, and targeted show marketing.
Funding Opportunities Take advantage of government support programs such as INQA coaching to advance your corporate strategies with financial assistance.
Digital excellence Comprehensive service management and professional IT security consulting enable you to optimize your business processes and protect your systems effectively. |
| Contact Now |
If you have any questions, please do not hesitate to contact us at +49 431 90881145 or by e-mail at store@railmarketresearch.com. |
|
|
|
Featured Research Projects | |
|
| We want to connect researchers and interested parties from industry and business and promote the exchange of research results.
If you are currently working on an exciting research project and would like to present it to a wider audience, please send us an e-mail. |
| Reach out now |
|
|
|
| | |
UIC Passenger Week Paris, France |
|
| | |
|
|
Intermodal South America São Paulo, Brazil |
|
|
|
World Travel Catering & Onboard Services Expo Hamburg, Germany |
|
|
|
Railway Diagnostic and Monitoring Conference Hamburg, Germany |
|
|
|
ERTMS 2026 Conference Valenciennes, France
|
|
|
|
Intermodal Asia Shanghai, China |
|
|
|
|
Inside RailMarketResearch | |
|
|
Receive the latest news on the following topics:
| Trends, news and expert contributions,
| upcoming events, | studies |
| Subscribe now! |
If you have any questions, please do not hesitate to contact us at +49 431 90881145 or by e-mail at store@railmarketresearch.com. |
|
|
|
|
