Better localization of trains with sophisticated multi-sensor-fusion supported by digital maps
Holistic condition monitoring including data from embedded wayside sensors and in-service trains
User-centered approach to Remote and Automatic Train Operation (RTO/ATO)
High Definition Simulation Environments to evaluate from innovative operational procedures up to necessary sensor capabilities to run Automated Train Operation efficiently
Accredited laboratory for conformity testing of ETCS, EULYNX and beyond
Focus: rail, digitalisation, automation, testing
At InnoTrans an interactive touch table demonstrates the solutions the researchers are currently developing. The demonstrator shows the following research topics of the DLR Institute of Transportation Systems:
Rail vehicle positioning with onboard sensors (RailPos)
Accurate and reliable position information is essential for digital and automated railway use cases. Computing it from onboard sensors remains a challenge. We contribute to the solution by developing map-supported multi-sensor fusion algorithms and open-source software for both real-time and post-processing use cases. Thanks to the modular structure, the algorithms are not tied to special hardware. They can also be used both offline and online applications.
Our approaches are being developed on real data from several application-driven projects. Current examples are the close collaboration with SBB (project EGNSS MATE, ETCS focus) or public transport operators in several European cities (project OnboardEU, focusing on condition monitoring and rail noise). We are also researching this topic in the RPS-Pro and R2DATO projects.
Further information can also be found in the two following publications, available via our Electronic Library elib:
CMTrain – Condition Monitoring with In-Service Trains
The maintenance of railway infrastructure requires precision and the highest quality, while the financial resources are limited. CMTrain offers the solution of condition monitoring during operation utilizing in-service trains. Sensors on regular rail vehicles enable real-time monitoring without disrupting operations. The advantage is track-accurate georeferencing and therefore precise positioning of railway defects. Data from different sources and systems can also be combined. CMTrain offers digital signal processing and machine learning: This can optain valuable status information about the track and the vehicle.
Complexity with physical cross-dependencies between infrastructure and/or asset components as well as train operation is one of the challenges in thoroughly monitoring the rail infrastructure. Simple models and single-sensor approaches are often not able to adequately capture the underlying causal relations in terms of fault propagation and data interdependence. By combining data-based multi-sensor fusion with causal modelling, Holistic Condition Monitoring offers tools for effectively discovering the root causes of failures instead of the symptoms only. It helps to digitalize and automate maintenance support as well as to provide better diagnostics. This will help to reduce costs and finally optimizes traffic and train operations.
RTO is considered an integral part of the migration strategy towards automated rail operations, because it forms a a fall-back layer in case of disruptions and an addition for non-centralized areas of operation, for example yards.
DLR has been involved in the development of RTO in various projects and consortiums since the very start. The researchers are developing requirements for RTO, a suitable workplace design and the necessary operational procedures, all taking human factors into account. They can rely on modular, particularly flexible and realistic simulators, many years of experience and a professional pool of test subjects.
The current rail infrastructure is working to capacity. However, the construction of new tracks is very expensive. Virtual coupling is an innovative solution to move more trains on the existing infrastructure. With virtual coupling trains no longer have to maintain the absolute braking distance, but can follow each other in relative braking and form a trainset. The necessary distance depends on speed, braking force, train length, as well as technical communication times and measurement inaccuracies.
The Institute of Transportation Systems examines the worthwile Use Case “slip coaching”. Slip coaching can reduce travel times while providing a maximum number of intermediate stops and connections at the same time – without having to change the infrastructure or build faster trains. Using Virtual Coupling Technology, multiple train units can run as one trainset that can split and stop independently from each other. The advantages are: Reduction of travel times of over 10%, doubled track capacity an more direct connections. See our demonstrator here: https://ngt.dlr.de
With OSTAR, the Institute of Transportation Systems offers a highly accurate simulation of vehicles and routes, including sensor technology. Validated sensor models and weather simulations are also integrated. If required, OSTAR also enables hardware/software-in-the-loop for automation logic or control and safety technology.
OSTAR can calculate the performance of a specific combination of track, train and sensor setup. It determines restrictive constraints and can evaluate measures to improve performance or safety. It can also evaluate offers during a tendering process, to provide an independent performance and safety evaluation of the tenders.
The system is currently used in several automotive projects and has been adapted to the railroad domain recently and is used in research projects.
RailSiTe® – Simulation and test laboratory for the railway sector
In the RailSiTe® (railway simulation and testing) laboratory, DLR has been researching new concepts and procedures for the faster commissioning of complex communication systems for rail transport for more than 10 years. It enables detailed technical and operational simulations as well as testing of railway management and signaling technology. The laboratory is capable of mapping the entire chain of the rail transport system in functional detail, including the trackside management and signaling technology, the interlocking, the track infrastructure and the air interface between track and train to the side of the vehicle. Railway operators benefit from its modular architecture, which makes it possible to integrate individual components of the management and signaling technology into the simulation both as software modules and as hardware components via real physical interfaces.
For ETCS and EULYNX tests, the laboratory is additionally flexibly accredited in accordance with DIN-ISO 17025. This means that even if there is a change in the version of the specifications, tests relevant to approval can be carried out immediately - without the necessity of renewed auditing by the Deutsche Akkreditierungsstelle (German accreditation body).
You can find more information on the RailSiTe®, it’s technical data, ongoing projects and collaborations here.
The RailSiTe® laboratory at the DLR site in Braunschweig.
When analyzing connected driver assistance systems, the institute has focused on a holistic view of the operational process. This not only takes into account the individual trajectory of a train in terms of its energy consumption and deviation from the timetable (punctuality, delays), but also the operational interaction with other trains regard to expected passenger volumes (passenger changeover times), capacity bottlenecks and peak load avoidance (energy provision).
In particular, the shifting of a timetable position to the next peak load interval results in economic savings potential (especially in the case of high local train volumes, e.g. within the Integral interval timetable), but with the tolerance of a corresponding delay. For more detailed information, please contact the Institute via the e-mail below.
You will find this exhibit and many research projects from the rail sector at the InnoTrans from 24 to 27 September 2024 in Berlin (Hall 2.2, Stand 440).
We are researching this topics in the departments:
