The Fraunhofer-Gesellschaft
The Fraunhofer-Gesellschaft, headquartered in Germany, is the world’s leading applied research organization. With its focus on developing key technologies that are vital for the future and enabling the commercial exploitation of this work by business and industry, Fraunhofer plays a central role in the innovation process. As a pioneer and catalyst for groundbreaking developments and scientific excellence, Fraunhofer helps shape society now and in the future. Founded in 1949, the Fraunhofer-Gesellschaft currently operates 75 institutes and research institutions throughout Germany. The majority of the organization’s 29,000 employees are qualified scientists and engineers, who work with an annual research budget of 2.8 billion euros. Of this sum, 2.4 billion euros are generated through contract research. The Fraunhofer Institute for High-Speed Dynamics, Ernst-Mach-Institut, EMI for short, is one of those institutes.
Profile of Fraunhofer EMI
Fraunhofer EMI recognizes its task in understanding the physics of high-speed, transient processes in order to derive solutions for applications in the industry. EMI adopts interdisciplinary and comprehensive material strategies to investigate crash, impact, and shock-wave phenomena by experiment and simulation.
The solutions elaborated at EMI by experimental, computer-based and analytical methods aim at improving the security and reliability of components and structures under dynamic loads. With the utilization of the most recent research results for technical applications, their efficiency is increased – conserving resources at the same time. Thereby, society profits from optimized systems in the areas of defense, security and resilience, automotive, space, aviation and sustainability.
The security of living environments and working spaces as well as the protection of urban environments and critical infrastructure are central topics in the business unit Security and Resilience. Through targeted risk management, solutions against extreme conditions such as explosions, impacts, fire, strong wind events and earthquakes can be elaborated. We provide safety, efficiency and robustness analyses of technical systems and develop sensor systems and software for safety and security applications.
Integration into the academic research landscape, on the one hand, as well as cross-linking and competent presence in the development departments of the relevant industry, on the other hand, are prerequisites for pursuing applied research in such an economical way as is common standard for Fraunhofer Institutes.
Three different work groups of EMI participate in the work in TeamAware: “Reliability of Learning Systems”, “Intelligent Data and X-Ray Analysis” and “Agent-Based Modeling”. Within their research scope the application and reliability of AI methods in the context of security applications is a main focus.
Role within the project TeamAware
With this expertise, Fraunhofer EMI was assigned WP Lead in WP10 “TeamAware Software AI Platform”. In preparation for WP10, EMI partakes in WP2 (“System Architecture Specifications and Design”) as Task Lead for T2.4 (“TeamAware Software Platform Architecture and Design”) to develop the first iteration for the platform software design documents in close relation to the other participating partners in TeamAware.
The work performed in T2.4 acts as a foundation for the AI software platform itself as well as it plays a major role in the later integration of the individual works from other WPs within TeamAware. Within the realization of this software platform during WP10, the main focus is on providing a reliable data source for the TeamAware ecosystem based on our database expertise. Also, the enhancement of individual sensor data by employing different methods of analytical and AI-based single and multi-sensor data fusion are also a major research topic in this. The final goal is the provision of the most relevant data to the operator at any given time while providing the operator with a powerful set of tools for decision support to plan and act on any emergency scenario. WP10 plays a major role in bringing the TeamAware data and benefits to the first responders.
To achieve the best possible integration of WP10 within the TeamAware ecosystem, Fraunhofer EMI also participates in WP2 for the system planning, WP9 (“Secure and Standardized Cloud network”) for the network and cloud systems, WP12 (“Integration and Test”) and WP13 (“Demonstration and Validation”).
Dr. Katharina Ross
Dr. Katharina Ross received her PhD (Dr. rer. nat.) at Friedrich Schiller University Jena in 2013 for the thesis: “Multi-scale Model for Transport Coefficients in Heterogeneous Fractured Media” in the field of computational hydrosystems in a joint project together with the Global Research for Safety (GRS) in Braunschweig for modeling the groundwater transport of radionuclides. Since 2016, she has worked for Fraunhofer as research associate and responsible person for the project “EDEN” within the department “Safety Technologies and Protectives Structures” and has focussed on modeling and evaluation of the possible damage effects on infrastructure after natural and man-made hazards and an efficient resilience management. Since 2019, she has been the manager of the group “Reliability of Learning Systems”, which focusses on AI-based systems and the reliability of its results to open the black box.
Jakob Stigler
Jakob Stigler has held a position as research associate at Fraunhofer EMI in the group “Intelligent Data and X-Ray Analysis” since 2020 and a MsC in Computer Science from the University of Freiburg with a specialization for neural-network-based AI. In 2013, he started at Fraunhofer EMI as student assistant and was involved in security projects like DURCHBLICK.
Dr. Corinna Köpke
Dr. Corinna Köpke studied geophysics and -informatics at the TU Bergakademie Freiberg, Germany, and received her master’s degree in collaboration with the Max Planck Institute for Solar System Research in Göttingen, Germany. In 2018, she received a PhD in geophysics from the University of Lausanne, Switzerland. Her work was focussed on the influence of modeling errors in inverse theory and uncertainty quantification. Afterwards, she spent one year at the DLR Institute for the Protection of Maritime Infrastructures in Bremerhaven, Germany, and joined Fraunhofer EMI in 2019. She focusses her work on resilience quantification and agent-based modeling for critical infrastructure protection and was involved in the EU project SATIE. As of recently, she fills the position of deputy group manager for “Agent-Based Modeling”.
Dr. Victoria Heusinger
Dr. Victoria Heusinger received her PhD (Dr. rer. nat.) at the University of Freiburg in 2019 for her work focussed on the optimization of computed tomography algorithms concerning the reduction of reconstruction artefacts, especially on undersampled data sets. The investigated iterative algebraic reconstruction method (ART) is very similar in its setup to the structure and learning process of neural networks, which is one of the reasons why, as of 2021, she manages the research group “Intelligent Data and X-Ray Analysis”, which focusses on AI application in the field of measurement data analysis (X-ray, radar, optical, data fusion), and especially the reduction of training data needed and the topic of “frugal learning”. Victoria has been at EMI since 2012 and participated in many projects in the area of security research, such as ECSIT, XP-Dite and DURCHBLICK.
Andreas Weber
Andreas Weber studied physics at Technical University Darmstadt where, during his master’s thesis at the Institute of Nuclear Physics, he specialized in machine learning. He implemented and investigated neural networks and their ability to solve complex tasks with the goal to improve the position resolution of a photon detector. Since 2020, he has been a research associate at Fraunhofer EMI in the group “Reliability of Learning Systems”, which focusses on AI-based systems and the reliability of their predictions.
