Dr. Tobias Brosch


I am a Machine Learning Specialist for Radar at Hensoldt Sensors GmbH in Ulm. This page maintains my former university web page. Please feel free to contact me via Linkedin.


  • Since 2019: Machine learning specialist for radar at Hensoldt Sensors GmbH
  • 2016-2019: Machine learning development specialist for autonomous driving and next generation multi-modal user interaction at BMW Car IT GmbH
  • 2016 (Jan-June): Postdoctoral researcher at Ulm University
  • 2015: Research intern IBM – Almaden (cognitive computing group)
  • 2010-2015: Phd student at Ulm University (studentship of “Graduate School Mathematical Analysis of Evolution, Information and Complexity”)
  • 2006-2010: Diploma Mathematician
  • 2006: Community Service
  • 2005: Abitur (Simpert Kraemer Gymnasium Krumbach)

Research Interests

  • Object classification/pattern recognition
  • Deep learning
  • Autonomous driving
  • Simulation
  • Driving strategy
  • GNSS positioning
  • Training simultaneous recurrent neural networks
  • Neuromorphic computing
  • Visual routines
  • Neural dynamics
  • Reinforcement learning
  • Machine learning
  • Asynchronous event based sensing
  • Computer vision

Work Experiences/Teaching/Certifications

  • 2017: Certified LeSS Practitioner
  • 2017: ISO 26262 based BMW Group Standard GS 95014 Training
  • 2017: Professional Scrum Master I (scrum.org)
  • 2016: ADTF Training
  • 2013: Work collaboration with Prof. Dr. Pieter Roelfsema at the Netherlands Institute of Neuroscience (1 month)
  • 2012-present: Sports instructor at  “Hochschulsport Ulm University
  • 2011-2014: Teaching assistant of “Natural Computation”, “Computer Vision 2”, “Information Processing in the Brain”, “Introduction to Computer Science” (Prof. Dr. H. Neumann, Prof. Dr. G. Palm)
  • 2010/11: Teaching assistant of “Object Oriented Programming with C++” (Dr. A. Borchert)
  • 2007–2009: Working as Tutor and typesetting two mathematical lecture notes of Prof. Dr. Spodarev in LaTeX.
  • 2008/2009: ”Studentische Hilfskraft” at the ”Institut für Angewandte Informationsverarbeitung” University Ulm. Creating web-services to access a java-application.
  • 30. Aug 04 to 3. Sep 04: Voluntary internship at AAI-Development-Services section MS2—elementary bioanalytical methods for quantitative determination of active pharmaceutical ingredients in biology matrices by using LC-MS respectively GC-MS.



  • Ufer, M., Mhasawade, V., Graef, R., Appel, H., Brosch, T. (2023) Micro-Doppler Based Deep Learning Approaches for Radar Applications. 24th International Radar Symposium (IRS).


  • Schwaiger, M., Kobold, J., Neumann, C., Brosch, T. (2022) Ultrafast Object Detection on High Resolution SAR Images. 23rd International Radar Symposium (IRS).


  • Neumann, C., Brosch, T. (2021) Automatic Target Recognition on High Resolution SAR Images with Deep Learning Domain Adaptation. 22nd International Radar Symposium (IRS). Pages 1-6.
  • Wrabel, A., Graef, R., Brosch, T. (2021) A Survey of Artificial Intelligence Approaches for Target Surveillance with Radar Sensors. IEEE Aerospace and Electronic Systems Magazine, 36(7):26-43.


  • Neumann, C., Brosch, T. (2020) Deep Learning Approach for Radar Applications. 21st International Radar Symposium (IRS). Pages 28-30



  • Brosch, T., Grünwedel, S. and Connette, C. (2018). Facets of the Usage of high-precision digital Maps for Driver Assistance and Autonomous Driving. VDI 2018.



  • Brosch, T. (2016). Kognitiver sequentieller Parallelismus: Von kanonischen neuronalen Schaltkreisen und dem Training rekurrenter neuronaler Netze für perzeptuelle Entscheidungsfindungen. In Hölldobler, S., et al. editors, Ausgezeichnete Informatikdissertationen 2015, pages 51–60. Lecture Notes in Informatics (LNI) – Dissertations, Volume D-16. ISBN: 978-3885799757
  • Jimeno Yepes, A., Tang, J., Saxena, S., Brosch, T. and Amir, A. Weighted Population Code for Low Power Neuromorphic Image Classification. Accepted for publication in Proc. IJCNN (IEEE), 2016.
  • C. Jarvers, T. Brosch, A. Brechmann, M. L. Woldeit, A. L. Schulz, F. W. Ohl, M. Lommerzheim, and H. Neumann. Reversal Learning in Humans and Gerbils: Dynamic Control Network Facilitates Learning. Frontiers in Neuroscience, 10(535):1–17, 2016
  • T. Brosch, S. Tschechne, and H. Neumann. Visual Processing in Cortical Architecture from Neuroscience to Neuromorphic Computing. In Brain–Inspired Computing 2015, volume 10087 of LNCS, pages 86–100. Springer, 2016.


  • Brosch, T., Tschechne, S. and Neumann, H. (2015). On Event-Based Optical Flow Detection. Frontiers in Neuroscience, 9(137):1–15.
  • T. Brosch. Cognitive Sequential Parallelism: From Canonical Neural Circuits to Training Recurrent Neural Networks in Perceptual Decision–Making. PhD thesis, Ulm University, 2015.
  • Brosch, T., Neumann, H. and Roelfsema, P. R. (2015). Reinforcement Learning of Linking and Tracing Contours in Recurrent Neural Networks. PLoS Computational Biology, 11(10):e1004489.
  • Brosch, T. and Neumann, H. (2015). Event-Based Optical Flow on Neuromorphic Hardware. In 9th EAI International Conference on Bio-inspired Information and Communications Technologies, pages 551-558. ICST.
  • Layher, G., Brosch, T. and Neumann, H. (2015). Towards a Mesoscopic-Level Canonical Circuit Definition for Visual Cortical Processing. In CMVC 2015, pages 543-550 .


  • Brosch, T. and Neumann, H. (2014). Interaction of Feedforward and Feedback Streams in Visual Cortex in a Firing–Rate Model of Columnar Computations. Neural Networks, 54:11–6.
  • Brosch, T., Roelfsema, P. and Neumann, H. (2014). Learning of New Perceptual Groupings – A Biologically Plausible Recurrent Neural Network Model that Learns Contour Integration. Journal of Vision (Abstract, VSS), 14(10):941
  • Brosch, T. and Neumann, H. (2014). Computing with a Canonical Neural Circuits Model with Pool Normalization and Modulating Feedback. Neural Computation, 26:12, pages 2735-89.
  • Tschechne, S. and Brosch, T. and Sailer, R. and von Egloffstein, N. and Abdul-Kreem, L. I. and Neumann, H. (2014). On Event-Based Motion Detection and Integration. In 8th International Conference on Bio-inspired Information and Communications Technologies, pages 298-305. ICST.


  • Brosch, T., Schwenker, F., and Neumann, H. (2013). Attention–Gated Reinforcement Learning in Neural Networks–A Unified View. In ICANN, volume 8131 of LNCS, pages 272–9. Springer.
  • Schels, M., Glodek, M., Meudt, S., Scherer, S., Schmidt, M., Layher, G., Tschechne, S., Brosch, T., Hrabal, D., Walter, S., Palm, G., Neumann, H., Traue, H., and Schwenker, F. (2013). Multi–Modal Classifier–Fusion for the Recognition of Emotions. In Rojc, M. and Campbell, N., editors, Coverbal Synchrony in Human–Machine Interaction, chapter 4, pages 73–98. CRC Press.


  • Brosch, T., Neumann, H. (2012). The Brain’s Sequential Parallelism: Perceptual Decision-Making and Early Sensory Responses. In ICONIP (Part II), volume 7664 of LNCS (pp. 41-50).
  • Brosch, T., Neumann, H. (2012). Perceptual Crowding in a Neural Model of Feedforward-Feedback Interactions. Journal of Vision (Abstract, VSS), 12(9):329.
  • Brosch, T., Neumann, H. (2012). The Combination of HMAX and HOGs in an Attention Guided Framework for Object Localization. ICPRAM (pp. 281-288).
  • Scherer, S., Glodek, M., Layher, G., Schels, M., Schmidt, M., Brosch, T., Tschechne, S., Schwenker, F., Neumann, H., and Palm, G. (2012). A Generic Framework for the Inference of User States in Human Computer Interaction. Journal of Multimodal User Interface, 6(3–4):117–41.


  • Layher, G., Tschechne, S., Scherer, S., Brosch, T., Curio, C., Neumann, H. (2011). Social Signal Processing in Companion Systems – Challenges Ahead. Proc. 41th Conference of the Gesellschaft für Informatik (Informatik’11). GI-Edition LNI, 239.
  • Glodek, M., Tschechne, S., Layher, G., Schels, M., Brosch, T., Scherer, S., Kächele, M., Schmidt, M., Neumann, H., Palm, G., Schwenker, S. (2011). Multiple Classifier Systems for the Classification of Audio-Visual Emotional States. Proc. 4th International Conference on Affective Computing and Intelligent Interaction (ACII’11). Springer LNCS 6975, 359-368.


  • May 2017: Digital Hackathon BMW (big data analysis)
  • July 2016: Promotionspreis of the Ulmer Universitätsgesellschaft
  • 2015: Nomination for GI-Dissertationspreis of the Gesellschaft für Informatik
  • November 2014: Invitation to IBM’s Cognitive Systems Colloquium in Almaden (on Truenorth’s architecture developed in DARPA funded SYNAPSE project)
  • June 2010: Achieved studentship of “Graduate School Mathematical Analysis of Evolution, Information and Complexity” (University of Ulm).
  • 2006: Participation at “Jugend forscht”—area mathematics/informatics— attained exceptional award of fundamental research.
  • 2005: Participation at the Regionalwettbewerb “Jugend forscht”— area physics.
  • Jan. 2005: Participation at the TU-Munich AbiTUMath-program (exponential functions on time-scales).
  • Sep. 2003 and 2004: Reached 2. round in Federal Republic of Germany selection process of 35. and 36. international physics olympiade.


T. Brosch, 2017. Zentralverriegelungssystem für ein Kraftfahrzeug, Patent DE 10 2017 215 413 A1, Sep 4, 2017.