Various approaches to autonomous driving in railways
https://doi.org/10.21683/1729-2646-2025-25-1-4-10
Abstract
In this paper, we analyse various technical solutions for autonomous driving. Depending on the role of an autonomous system, different safety integrity levels may be required. We examine the three primary architectures. The first one is simply a support system that only requires a basic integrity, rather than a level. The second one is a simple replacement of the driver, which corresponds to SIL 1 up to SIL 2. The third architecture is an integration of ATO into a safe train protection system, which corresponds to SIL 4.
About the Authors
Hendrik SchäbeGermany
Hendrik Schäbe, Doctor of natural sciences, Main specialist on RAMS,
Cologne.
Igor B. Shubinsky
Russian Federation
Igor B. Shubinsky, Professor, Doctor of Engineering, Chief Expert,
Moscow.
Efim N. Rozenberg
Russian Federation
Efim N. Rozenberg, Professor, Doctor of Engineering, First Deputy Director General,
Moscow.
References
1. Hoekstra S., Middelkoop D., de Vries D. et al. ATO Verkenning Betuweroute Eindrapportage, 1.0, 21.7.2020, ProRail.
2. Lieskovsky A., Mysliviec I., Zemlicka M. Automatic Train Operation: History and Open Questions. In: Proceedings of VEHITS 2020 – 6th International Conference on Vehicle Technology and Intelligent Transport Systems; 2020. Pp. 260-267.
3. Schäbe H. Autonomes Fahren auf der Schiene und der Straße – Erfahrungen nutzbar machen für die Schiene. Fachtagung Autonomes Fahren und Betriebshofautomatisierung im Schienenverkehr (30.09.2021 – 01.10.2021). Potsdam; 2021.
4. ERA, TSI revision 2022. Digital Rail and Green Freight. Changes proposed to the CCS TSI. Version 1.0, 2022-03-18.
5. Tigadi A., Gujanatti R., Ginchi A., Advanced Driver Assistance Systems. International Journal of Engineering Research and General Science 2016;4(3):151-158.
6. Schäbe H., Wigger P. Experience with SIL Allocation in Railway Applications. In: Proceedings of the 4th International Symposium “Programmable Electronic Systems in Safety Related Applications”, TÜV, 3-4 May 2000. Cologne; 2000.
7. Braband J., Evers B., Kinas M. et al. Risikoakzeptanzkriterien für das automatisierte Fahren auf der Schiene. Berichte des Deutschen Zentrums für Schienenverkehrsforschung, Bericht 40 (2023). DOI: 10.48755/dzsf.230008.01.
8. Braband J., Lindner L., Rexin F. Risk analysis for obstacle detection in automated driving. Signal & Datacommunication 2023;3:12-20.
9. Adebahr F., Schäbe H. Wahrscheinlichkeit menschlicher Fehler und Ausfallraten technischer ATO-systeme. Signal und Draht 2023;3(115):21-29.
10. DIN VDE V 0831-103:2020-09 Electric signalling systems for railways – Part 103: Identification of safety requirements for technical functions in railway signalling; 2020.
11. Braband J., Schäbe H. Risikoanalyse für das Automatisierte Fahren – Validierung und Erkenntnisse. Signal und Draht 2023;4(115):6-12.
12. Shubinsky I.B., Shaebe H., Rozenberg E.N. On the safety assessment of an automatic train operation system. Dependability 2021;4:31-37.
13. MODSAFE, WP2–D2.2 Consistency analysis and Final hazard Analysis. TU Dresden, DEL_D2.2_TUD_ WP2_100430_V1.1, 2010-04-30.
14. MODSAFE, WP4–D4.2 Analysis of Safety requirements for MODSafe Continuous Safety Measures and Functions. DEL_4.2_UITP_WP4_11021_V2.0, 2011-01-21
15. MODSAFE, WP4.3–D4.3 Analysis of On Demand Functions. DE_D4.3 3_UITP_WP4_120220_V 1.3, 202102-15.
16. Vom Hövel R., Braband J., Schäbe H. The probability of failure on demand – the why and the how. In: Proc. Int. Conf. on Computer Safety, SafeComp 2009, LNCS 5775. Springer: Berlin-Heidelberg; 2009. Pp. 46-54.
17. Smaki B. Artificial Intelligence applications for railway signalling. KTH Stockholm; 2021.
Review
For citations:
Schäbe H., Shubinsky I.B., Rozenberg E.N. Various approaches to autonomous driving in railways. Dependability. 2025;25(1):4-10. (In Russ.) https://doi.org/10.21683/1729-2646-2025-25-1-4-10