Ongoing projects

OHMetaRunner

Autonomously driving model vehicles based on explainable meta learning classifiers

The automotive industry has been called on by political and social requirements to develop transparency for its ML algorithms. Specific and real experiences in meta learning will play an important role in this.

“OHMeta Runner” will provide these types of experiences for the first time by using an innovative robotic construction of an autonomous model vehicle in a modelling scale (ca. 1:8) and an obstacle course. The focus of the research is on exploring the implementation of meta learning technology for autonomous driving and the analysis of the applicability of explainable AI approaches to this technology. This means the machine learning decisions will be traceable and transparent, and on an architectural level, an abstract security-risk analysis can be executed.

Our team is developing a versatile-purpose robotic construction (model vehicle) that is robust, has all required sensors and actuators, and meets the high performance and efficiency requirements of ML algorithms. A subsequent migration to Adaptive AUTOSAR is certainly an aim for follow-on projects and it is, in principle, primed for this due to the conformity with EAST-ADL on the architectural level.

Creating this robot construction has resulted in a fruitful inter-Faculty, interdisciplinary platform that will generate novel ideas for the area of software engineering and machine learning.

Safety and Security in Machine Learning for Autonomous Driving

Consortial doctorate in Range Safety und Security Engineering

Research collaboration with the Division of Computer Science and Software Engineering at Mälardalen University, Sweden

Research content:

  • Concept for extensions of the SAM-meta model for safe and secure vehicle software architectures
  • Early detection of concept drift in ML-based algorithms using statistical distance calculations
  • Characterization of data sets

Individuals involved in the doctoral process:

  • Matthias Bergler (doctoral candidate, the Ohm)
  • Prof. Ramin Tavakoli Kolagari (UAS doctoral supervisor, the Ohm)
  • Prof. Dr.  Kristina Lundqvist (doctoral supervisor, Mälardalen University, Sweden)

 

Ongoing up to and including March 2023

Design, set-up, and operative testing of a fully automatic shunting engine (“Vollautomatische Rangierlok”, VAL) in cooperation with the Deutsche Bahn (DB Cargo).

In the VAL (fully automated shunting locomotive) project, a retrofit module for a shunting locomotive is being developed and brought to application maturity. The module allows for fully automated shunting without the presence of a locomotive shunting driver. Individual components from other branches of industry are modified and combined to form an innovative, novel system. In addition, a new digital interface will be created between the infrastructure and the locomotive, which will provide an adequate replacement for the current verbal communication. A new computer component will also be integrated into the signal box, which will be used to transmit driving commands to the locomotive. The result is a product that is ready for application, that fulfils all the necessary safety-critical requirements, and that enables a series release with market launch.

For the validation and evaluation of the functions that are to be automated in order to drive the fully automated locomotive, specific tests, which include relevant driving manoeuvres under environmental influences, are necessary in order to compare the reaction of the overall system to the respective situation against what is expected of it. As a result, the Ohm is focusing on developing methods and criteria for testing functional requirements from the point of view of perception, localisation, classification, and the resulting driving and braking decision.

For the performance of the tests, two test environments will be created in parallel, within which a systematic and consistent test procedure will be developed and applied. A laboratory environment and a characteristic real-world environment in the field will be created as test environments, in which the system’s reactions to defined disturbance and input variables will be evaluated on the basis of defined criteria. Traditional software verification methods such as regression testing are used, together with verification methods for software with high AI (artificial intelligence) content, such as Adversarial Examples/Attacks or Explainable Artificial Intelligence techniques.

The creation of test environments and scientific tests that are independent of the development of the system leads to increased system security, since the requirements cannot be described precisely due to the large number of possible situations, but still have to be evaluated according to well-defined criteria.

The project is funded by the Federal Ministry of Transport and Digital Infrastructure.

Collaboration with: DB Cargo AG, University of Bamberg — Cognitive Systems

Project duration: December 2020 to December 2024

Completed projects

FORMUS³IC

“Multi-Core safe and software-intensive Systems Improvement Community”
  • Research content: Overcoming the challenges facing the automotive and aviation industry arising as a result of heterogeneous multi-core architectures
  • Research association of the Laboratory for Safe and Secure Systems working group (LaS³)
  • Academic project partners OTH Regensburg, TH Ingolstadt, OTH Amberg-Weiden, HS Munich, FAU Erlangen-Nuremberg
  • Industrial project partners Audi AG, Continental Automotive GmbH (Powertrain), AIRBUS Defence and Space GmbH, Elektrobit Automotive GmbH, Infineon Technologies AG, iNTENCE automotive electronics GmbH, Timing-Architects Embedded Systems GmbH, XKrug GmbH
  • Homepage: https://formus3ic.de

 

Completed December 2018

Funded by the Bavarian Research Foundation

IT security for autonomous and networked driving

Consortial doctorate in automotive IT security

Research partnership with the Chair for Applied Cryptography at Friedrich-Alexander-University Erlangen-Nuremberg

Research content:

  • Designing safe and secure vehicle software architectures by expanding the EAST-ADL
  • Investigating safety-critical vulnerabilities through penetration tests
  • Designing a safe platform as a framework

Individuals involved in the doctoral process:

  • Markus Zoppelt (doctoral candidate, the Ohm)
  • Prof. Ramin Tavakoli Kolagari (UAS doctoral supervisor, the Ohm)
  • Prof. Dominique Schröder (university doctoral supervisor, CHAAC, Friedrich-Alexander-University Erlangen-Nuremberg)

 

Degree completed in March 2021

Funded by the Zentrum Digitalisierung.Bayern

MAENAD

Model-based Analysis & Engineering of Novel Architectures for Dependable Electric Vehicles

http://maenad.eu

 

Project duration: Sept 2010 – Aug 2013

 

Research associates: Prof. Ramin Tavakoli Kolagari, Tobias Wägemann

 

Publications:

Sara Tucci-Piergiovanni, De-Jiu Chen, Chokri Mraidha, Henrik Lönn, Nidhal Mahmud, Mark-Oliver Reiser, Ramin Tavakoli Kolagari, Nataliya Yakymets, Renato Librino, Sandra Torchiaro, Agnes Lanusse: Model-Based Analysis and Engineering of Automotive Architectures with EAST-ADL, Chapter 10 of the Handbook of Research on Embedded Systems Design, edited by Alessandra Bagnato, Leandro Soares Indrusiak, Imran Rafiq Quadri and Matteo Rossi, IGI Global, Hershey, August 2014. pp. 242–282.

 

Hans Blom, Henrik Lönn, Frank Hagl, Yiannis Papadopoulos, Mark-Oliver Reiser, Carl-Johan Sjöstedt, De-Jiu Chen, Fulvio Tagliabò, Sandra Torchiaro, Sara Tucci, Ramin Tavakoli Kolagari: EAST-ADL: An Architecture Description Language for Automotive Software-Intensive Systems, Chapter 23 of Embedded Computing Systems: Applications, Optimization, and Advanced Designedited by Mohamed Khalgui, Olfa Mosbahi and Antonio Valentini, IGI Global, Hershey, April 2013. pp. 456–470.

 

Philippe Cuenot, Patrik Frey, Rolf Johansson, Henrik Lönn, Yiannis Papadopoulos, Mark-Oliver Reiser, Anders Sandberg, David Servat, Ramin Tavakoli Kolagari, Martin Törngren, Matthias Weber (invited paper, peer reviewed): The EAST-ADL Architecture Description Language for Automotive Embedded Software, Invited Chapter in the LNCS Volume on Model-Based Engineering of Embedded Real-Time Systems (MBEERTS), Lecture Notes in Computer Science, edited by Holger Giese, Bernard Rumpe, Bernard Schätz, International Dagstuhl Workshop, Dagstuhl Castle, Germany, November 4-9, 2007, Revised Selected Papers, 2010. pp. 297–307.

 

Eric Armengaud, Andreas Baumgart, Matthias Biehl, De-Jiu Chen, Gerhard Griessnig, Christian Hein, Tom Ritter, Ramin Tavakoli Kolagari, Markus Zoier: Model-based Toolchain for the Efficient Development of Safety-relevant Automotive Embedded Systems, Proceedings of the SAE 2011 World Congress & Exhibition, Detroit, April 2011.

 

Anders Sandberg, De-Jiu Chen, Henrik Lönn, Rolf Johansson, Lei Feng, Martin Törngren, Sandra Torchiaro, Ramin Tavakoli Kolagari, Andreas Abele: Model-based Safety Engineering of Interdependent Functions in Automotive Vehicles Using EAST-ADL2, Proceedings of the 29th International Conference on Computer Safety, Reliability and Security (SAFECOMP’10), Vienna, September 2010. pp. 332–346.

 

 Philippe Cuenot, Patrick Frey, Rolf Johansson, Henrik Lönn, Yiannis Papadopoulos, Mark-Oliver Reiser, Anders Sandberg, David Servat, Ramin Tavakoli Kolagari, Martin Törngren, Matthias Weber: The EAST-ADL Architecture Description Language for Automotive Embedded Software, Chapter 11 of Model-Based Engineering of Embedded Real-Time Systems, International Dagstuhl Workshop, Dagstuhl Castle, Germany, November 4-9, 2007, Revised Selected Papersedited by Holger Giese, Gabor Karsai, Edward Lee, Bernhard Rumpe, Bernhard Schätz, Lecture Notes in Computer Science 6100, Springer, Berlin, 2010. pp. 297–308.

 

Hans Blom, Henrik Lönn, Ramin Tavakoli Kolagari: On Tooling strategy for the EAST-ADL language, Proceedings of the 1st Workshop on Hands-on Platforms and tools for model-based engineering of Embedded Systems (HoPES’10), in conjunction with ECMFA 2010, Paris, June 2010. pp. 19–26.

 

Hans Blom, Henrik Lönn, Frank Hagl, Yiannis Papadopoulos, Mark-Oliver Reiser, Carl-Johan Sjöstedt, De-Jiu Chen, Ramin Tavakoli Kolagari: EAST-ADL—An Architecture Description Language for Automotive Software-Intensive Systems, White Paper, Version 2.1.12, 2012.

OhmRunner

Car2X communication is technology that aims to enable the exchange of data, for example hazard sources, obstacles, traffic flow or the general vehicle status, with other road users or the infrastructure. The aim of the “Ohm Runner” project, supported by the Staedtler Foundation, is to investigate and assess how this communication is established, the various applications in this area, and possible measures to be taken to secure the integrity of the data transferred in this way. The project seeks to teach a model vehicle how to exchange information with its environment and interpret that information using vehicle standards and technologies such as AUTOSAR. This requires potential application scenarios to be defined, implemented, and assessed as regards their practicability and added value. The information generated here may become an important component of smart city and smart grid concepts and help not only to influence traffic management in real time but also to predict and even influence certain occurrences using skilful assessments of these large data quantities.

Work performed by Christian Stahl and Johannes Rösler.

Completed September 2017

Funded by the Staedtler Foundation

PetS³

Due to increasing networking, cyber attacks are a growing threat to many application areas, including networked driving or smart metering. These gateway-based systems require joint concepts of functional safety and IT security. The aim of the research project is to research attacks on the IT security of system architectures used for networked cyber-critical systems as the interaction between functional safety and IT security in this context is still largely unknown. The central purpose of the research is to comprehensively analyse the security technology, the gateway structures, and the corresponding system development processes. Requirements for system architectures, security technology, and test processes, as well as for a maturity model will be derived using vulnerabilities and threat analyses, together with penetration tests and architectural analyses. A catalogue of vulnerabilities and threats will be developed to form the basis for an automated analysis of functional safety and IT security and will thereby make a core contribution to the risk analysis of cyber-critical systems.

Cooperation partners: Ostbayerische Technische Hochschule Regensburg, sepp.med GmbH, eMundo GmbH, iNTENCE automotive electronic GmbH, BFFT Gesellschaft für Fahrzeugtechnik mbH

SmartOSE

In order to protect danger zones from the consequences of sinkholes or other georisks, safety devices containing integrated geoplastics are used. These building products, which are predominantly manufactured by textile engineering, are highly elastic and can be used permanently and cheaply to reinforce the ground. They are also used as a protective measure against the effects of contaminated soils that pose a risk to the environment or erosion, as well as to seal off and drain landfills, waterways or dams and dykes. 

The research project “Optical sensors for monitoring earth structures”, led by Prof. Rainer Engelbrecht, aims to develop and investigate innovative and cost-effective sensors on the basis of polymer optical fibres (POF) to monitor the structure of geoplastics.

Further information can be found on the project homepage.

Our publications

2023

Bergler, Matthias, and Ramin Tavakoli Kolagari. "Automotive Software Security Engineering based on the ISO 21434." Proceedings of the 2023 5th World Symposium on Software Engineering. 2023.

2021

Bergler, Matthias; Tolvanen, Juha-Pekka; Zoppelt, Markus; Tavakoli Kolagari, Ramin (2021). Social engineering exploits in automotive software security: modeling human-targeted attacks with SAM. Proceedings of the 31st European Safety and Reliability Conference | 19-23 September 2021, Angers, France.

2020

Brost, J., Egger, C., Lai, R. W. F., Schmid, F., Schröder, D., & Zoppelt, M. (2020). Threshold Password-Hardened Encryption Services. Proceedings of the 2020 ACM SIGSAC Conference on Computer and Communications Security, 409-424.https://doi.org/10.1145/3372297.3417266

Aslansefat, K., Sorokos, I., Whiting, D., Tavakoli Kolagari, R., & Papadopoulos, Y. (2020). SafeML: Safety monitoring of machine learning classifiers through statistical difference measure. IMBSA 2020. Lecture notes in computer science, vol 12297, 197–211.https://doi.org/10.1007/978-3-030-58920-2_13

Auernhammer, Katja; Freiling, Felix; Tavakoli Kolagari, R. (2020). Efficient Black Box Search for Advesarial Examples Using Relevance Masks. DYnamic and Novel Advances in Machine Learning and Intelligent Cyber Security (DYNAMICS) workshop.

2019

Zoppelt, M., & Tavakoli Kolagari, R. (2019). Reaching Grey Havens: Industrial Automotive Security Modeling with SAM. International journal on advances in security, 12, 223-235.https://www.iariajournals.org/security/sec_v12_n34_2019_paged.pdf

Beuche, D., Birk, A., Dreier, H., Fleischmann, A., Galle, H., Heller, G., Janzen, D., John, I., Tavakoli Kolagari, R., von der Maßen, T., Wolfram, A., Omasreiter, H., Tavakoli Kolagari, R., Zoppelt, M., Tavakoli Kolagari, R., Mark-Oliver Reiser, Tavakoli Kolagari, R., Weber, M., van Wagensveld, R., … Mader, R. (2019). UnCle SAM : Modeling Cloud Attacks with the Automotive Security Abstraction Model. In Quan Chen Zhiyi Huang & P. Balaji (Eds.), CLOUD COMPUTING 2019, The Tenth International Conference on Cloud Computing, GRIDs, and Virtualization (Vol. 11681, Issue IEEE Computer Society, Los Alamitos, pp. 67–72). Springer, Berlin. doi.org/10.1007/978-3-540-75698-9_9

Zoppelt, M., & Tavakoli Kolagari, R. (2019). UnCle SAM : Modeling Cloud Attacks with the Automotive Security Abstraction Model. Proceedings of the Tenth International Conference on Cloud Computing, GRIDs, and Virtualization, 67-72.

Auernhammer, K., Tavakoli Kolagari, R., Zoppelt, M., & Tavakoli Kolagari, R. (2019). Attacks on machine learning: Lurking danger for accountability. CEUR workshop proceedings, 2301, 9.http://ceur-ws.org/Vol-2301/paper_2.pdf

Tavakoli Kolagari, R. (2019). Modeling Software Systems Security with SAM and EAST-ADL. doi.org/10.1007/978-3-030-32872-6

Zoppelt, M., & Tavakoli Kolagari, R. (2019). What Today’s Serious Cyber Attacks on Cars Tell Us: Consequences for Automotive Security and Dependability. International Symposium on Model-Based Safety and Assessment, 270-285.https://doi.org/10.1007/978-3-030-32872-6_18

Auernhammer, K., Tavakoli Kolagari, R., & Zoppelt, M. (2019). Attacks on machine learning: Lurking Danger for Accountability. Proceedings of the AAAI workshop on artificial intelligence safety, co-located with the Thirty-Third AAAI Conference on Artificial Intelligence (AAAI 2019), 9-18.

Wägemann, T., Tavakoli Kolagari, R., & Schmid, K. (2019). ADOOPLA - Combining Product-Line- and Product-Level Criteria in Multi-objective Optimization of Product Line Architectures. In I. P. Bures T. Duchien L. (Ed.), Lecture Notes in Computer Science (Vol. 11681). Springer, Cham. doi.org/10.1007/978-3-030-29983-5_9

van Wagensveld, R., Wägemann, T., Mader, R., Tavakoli Kolagari, R., & Margull, U. (2019). Evaluation and modeling of the supercore parallelization pattern in automotive real-time systems. Parallel Computing, 81, 122–130. doi.org/10.1016/j.parco.2018.12.004

Wägemann, T., Tavakoli Kolagari, R., & Schmid, K. (2019, March). Exploring Automotive Stakeholder Requirements for Architecture Optimization Support. Proceedings of the International Conference on Software Architecture Companion(ICSA-C). doi.org/10.1109/ICSA-C.2019.00015

Auernhammer, K., Tavakoli Kolagari, R., & Zoppelt, M. (2019). Attacks on machine learning: Lurking danger for accountability. CEUR workshop proceedings, 2301, 9.http://ceur-ws.org/Vol-2301/paper_2.pdf

Zoppelt, M., & Tavakoli Kolagari, R. (2019). What today’s serious cyber attacks on cars tell us: Consequences for automotive security and dependability. In M. Papadopoulos, Yiannis and Aslansefat, Koorosh and Katsaros, Panagiotis and Bozzano (Ed.), Lecture notes in computer science (including subseries Lecture notes in artificial intelligence and Lecture notes in bioinformatics): Vol. 11842 LNCS (pp. 270–285). Springer International Publishing. doi.org/10.1007/978-3-030-32872-6_18

Zoppelt, M., & Tavakoli Kolagari, R. (2019). UnCle SAM : Modeling Cloud Attacks with the Automotive Security Abstraction Model. CLOUD COMPUTING 2019, The Tenth International Conference on Cloud Computing, GRIDs, and Virtualization, 67–72.

Zoppelt, M., & Tavakoli Kolagari, R. (2019). SAM: A Security Abstraction Model for Automotive Software Systems. In Lecture notes in computer science (including subseries Lecture notes in artificial intelligence and Lecture notes in bioinformatics): Vol. 11552 LNCS (pp. 59–74). Springer. doi.org/10.1007/978-3-030-16874-2_5

2018

van Wagensveld, R., Wägemann, T., Hehenkamp, N., Tavakoli Kolagari, R., Margull, U., & Mader, R. (2018). Intra-Task Parallelism in Automotive Real-Time Systems. In Quan Chen Zhiyi Huang & P. Balaji (Eds.), Proceedings of the 9th International Workshop on Programming Models and Applications for Multicores and Manycores 2018 (PMAM’18) (pp. 61–70). ACM, New York. doi.org/10.1145/3178442.3178449

Zoppelt, M., & Tavakoli Kolagari, R. (2018). SAM: A Security Abstraction Model for Automotive Software Systems. In S. A. E. Y. G.-A. J. Hamid B. Gallina B. (Ed.), Security and Safety Interplay of Intelligent Software Systems (Vol. 11552, pp. 59–74). Springer, Cham. doi.org/10.1007/978-3-030-16874-2_5

Wägemann, T., Tavakoli Kolagari, R., & Schmid, K. (2018). Optimal Product Line Architectures for the Automotive Industry. In Schaefer I., D. Karagiannis, A. Vogelsang, D. Méndez, & C. Seidl (Eds.),.), Modellierung 2018 (pp. 119–134). Gesellschaft für Informatik. dl.gi.de/handle/20.500.12116/14962

2017

Stahl, C., & Tavakoli Kolagari, R. (2017). Strategies to balance grid load using communication interfaces in electrical vehicles. In J. Mottok, M. Reichenberger, & C. Klippel (Eds.), Applied Research Conference 2017. (ARC 2017).

Rösler, J., & Tavakoli Kolagari, R. (2017). Ohm runner---Establishment of a communication from infrastructure to vehicle level for dynamic transmission of information. In J. Mottok, M. Reichenberger, & C. Klippel (Eds.), Applied Research Conference 2017. (ARC 2017).

2016

Tavakoli Kolagari, R., Chen, D., Lanusse, A., Librino, R., Lönn, H., Mahmud, N., Mraidha, C., Reiser, M.-O., Torchiaro, S., Tucci-Piergiovanni, S., Wägemann, T., & Yakymets, N. (2016). Model-based Analysis & Engineering of Automotive Architectures with EAST-ADL: Revisited. International Journal of Conceptual Structures and Smart Applications (IJCSSA) 3 (2). doi.org/10.4018/IJCSSA.2015070103

Wägemann, T., Langer, T., Mottok, J., Osinski, L., Stappert, F., & Tavakoli Kolagari, R. (2016). Models for Dependable Heterogenous Multi- and Many-Core System Software Design Revisited. Proceedings of the FORMUS3IC workshop of multi-core safe and software-intensive systems improvement community (ARCS 2016), 1-8.https://ieeexplore.ieee.org/document/7499190

Tavakoli Kolagari, R., & Auernhammer, K. (2016, July). Utilization and Effect of Narrative Art in Advanced Software Engineering Education. Proceedings of the European Conference Software Engineering Education 2016.

Blom, H., Chen, D.-J., Kaijser, H., Lönn, H., Papadopoulos, Y., Reiser, M.-O., Tavakoli Kolagari, R., & Tucci, S. (2016). EAST-ADL---An Architecture Description Language for Automotive Software-intensive Systems in the Light of Recent use and Research. International Journal of System Dynamics Applications (IJSDA).https://doi.org/10.4018/IJSDA.2016070101