News

Publications

Jan Gruteser, Jan Roßbach, Fabian Vu, Michael Leuschel. Using Formal Models, Safety Shields and Certified Control to Validate AI-Based Train Systems. FMAS 2024, to appear.


Hans-Werner Wiesbrock, Juergen Grossmann. Probabilistically Extended Ontologies: A Basis for Systematic Testing of ML-Based Systems. SAE Technical Paper 2024-01-3002, 2024. The final authenticated version is available online at https://doi.org/10.4271/2024-01-3002


Hans-Werner Wiesbrock, Juergen Grossmann. Outline of an Independent Systematic Blackbox Test for ML Systems. The 6th IEEE International Conference on Artificial Intelligence Testing (IEEE AITEST 2024), 07/2024, Shanghai, China. The final authenticated version is available online at https://doi.org/10.1109/AITest62860.2024.00009


Nicolas Grube, Mozhdeh Massah, Michael Tebbe, Paul Wancura, Hans-Werner Wiesbrock, Juergen Grossmann, Sami Kharma. On a systematic test of ML-based systems: Experiments on test statistics. The 6th IEEE International Conference on Artificial Intelligence Testing (IEEE AITEST 2024), 07/2024, Shanghai, China. The final authenticated version is available online at https://doi.org/10.1109/AITest62860.2024.00010


Jan Roßbach, Olivier De Candido, Achmed Hammam, Michael Leuschel. (2024). Evaluating AI-Based Components in Autonomous Railway Systems. In: Hotho, A., Rudolph, S. (eds) KI 2024: Advances in Artificial Intelligence. KI 2024. Lecture Notes in Computer Science(), vol 14992 . Springer, Cham. The final authenticated version is available online at https://doi.org/10.1007/978-3-031-70893-0_14


Jan Gruteser, Michael Leuschel. Validation of RailML Using ProB. In Proceedings ICECCS 2024, LNCS, 14784, Springer, 245-256, 2024. The final authenticated version is available online at https://doi.org/10.1007/978-3-031-66456-4_13


Fabian Vu, Jannik Dunkelau, Michael Leuschel. Validation of Reinforcement Learning Agents and Safety Shields with ProB. In Proceedings NFM 2024, LNCS, 14627, Springer, 279-297, 2024. The final authenticated version is available online at https://doi.org/10.1007/978-3-031-60698-4_16


Fabian Vu, Christopher Happe, Michael Leuschel. Generating interactive documents for domain-specific validation of formal models. In International Journal on Software Tools for Technology Transfer, 26, Springer, 147-168, 2024. The final authenticated version is available online at https://doi.org/10.1007/s10009-024-00739-0


Jan Gruteser, David Geleßus, Michael Leuschel, Jan Roßbach and Fabian Vu. A Formal Model of Train Control with AI-based Obstacle Detection. Proceedings RSSRail 2023. LNCS. 14198, Springer, 128-145, 2023. The final authenticated version is available online at https://doi.org/10.1007/978-3-031-43366-5_8(HHU)


G. Hemzal, T. Strobel, J. Großmann, M. Leuschel, D. Knoblauch, M. Kucheiko, N. Grube, R. Krajewski: KI-LOK – Ein Verbundprojekt über Prüfverfahren für KI-basierte Komponenten im Eisenbahnbetrieb; Signal + Draht 04/ 2023


Grossmann, J. et al. (2023). Test and Training Data Generation for Object Recognition in the Railway Domain. In: Masci, P., Bernardeschi, C., Graziani, P., Koddenbrock, M., Palmieri, M. (eds) Software Engineering and Formal Methods. SEFM 2022 Collocated Workshops. SEFM 2022. Lecture Notes in Computer Science, vol 13765. Springer, Cham. (Fraunhofer FOKUS, ITPower Solutions); Verfügbar unter: https://link.springer.com/chapter/10.1007/978-3-031-26236-4_1


Klemenc, Jona, and Holger Trittenbach. „Selecting Models based on the Risk of Damage Caused by Adversarial Attacks.“ arXiv preprint arXiv:2301.12151 (2023) (neurocat); Available under: https://arxiv.org/pdf/2301.12151v1.pd


Jan Roßbach, Michael Leuschel. Certified Control for Train Sign Classification. In Proceedings FMAS 2023, EPTCS, 395, 69–76, 2023. The final authenticated version is available online at https://doi.org/10.4204/EPTCS.395.5


Juergen Großmann, Nicolas Grube, Sami Kharma, Dorian Knoblauch, Roman Krajewski, Mariia Kucheiko, Hans-Werner Wiesbrock. Test- und Trainingsdatengenerierung für die Objekterkennung im Bahnbereich. AI4EA 2022, Berlin, Germany.


Fabian Vu, Christopher Happe, Michael Leuschel. Generating Domain-Specific Interactive Validation Documents, FMICS 2022, LNCS, 13487, Springer, 32-49, 2022. The final authenticated version is available online at https://doi.org/10.1007/978-3-031-15008-1_4


G. Hemzal, T. Strobel, J. Großmann, B.-H. Schlingloff, M. Leuschel, S. Sadeghipour, J. Firnkorn: KI-LOK – Ein Verbundprojekt über Prüfverfahren für KI-basierte Komponenten im Eisenbahnbetrieb; Signal + Draht 10/ 2021 


11.08.2022

KI-LOK Project Report

The report provides information on the goals and strategies of the KI-LOK project for automated testing of AI-based perception systems using a tabular breakdown.

Er steht kostenlos zum Download zur Verfügung.