fortiss Research

fortiss Research

From fundamental research to marketable prototypes

fortiss has established itself in the world's top research on central topics of Software & Systems Engineering, AI Engineering and IoT Engineering and is considered a recognized partner for demanding questions in software development and Artificial Intelligence.

The institute develops and operates high-performance software with reliable functionality, performance, resilience, persistence, security and maintainability. The special focus is on the integration of model-driven software development with data-driven programming of AI, for the controllable development of a new generation of increasingly autonomous and also decentralized software systems.

    Main research areas

    Software & Systems Engineering

    How can software-intensive cyber-physical systems (CPS)/IoT be developed in a controllable way?


    The classical methods of model-based software and system development are extended and integrated with new development methods for data-driven applications. Until now, the focus has been on ensuring the reliability and security of relatively small, centralized and automated systems operating in predictable environments. Current and future requirements are increasingly shifting to ensuring trustworthiness of larger, dynamically networked, self-learning and constantly evolving, often autonomously acting systems. Current focus topics include the structured development of trustworthy autonomous systems, the analysis and optimization of software and system architectures, software engineering for data-driven applications, the use of AI methods in software engineering, and the assurance and certification of large software systems, among other research priorities.

    AI Engineering

    How can increasingly autonomous and mission-critical CPS/IoT be controllably developed with learning-based AI components?


    Despite technological advances that have led to the proliferation of AI-based and increasingly autonomous systems, the question of the level of trust that can be placed in these learning-based software systems remains. A new generation of robust AI technology that makes timely and confident decisions in uncertain, unpredictable environments is therefore needed for a wide range of applications. Their results are comprehensible, explainable, and resistant to erroneous input and targeted attacks. In addition, they can process ever-increasing amounts of data, but also increasingly derive useful insights from small data sets without making significant compromises to confidentiality and privacy. When developing and operating AI-powered software systems, it is important to take an engineering approach to ensure that AI technology can be used in mission-critical and security-relevant applications in the future.

    IoT Engineering

    How can software platforms for trusted, decentralized services be developed as enablers of a new generation of products?
     

    Prerequisites for a flexible, software-based infrastructure that adapts and optimizes itself as needed are the deep embedding of sensor, computing and communication capabilities in existing systems and the penetration of traditional, physical infrastructures. The foundation for smart infrastructures and the decentralized and increasingly data-based services that belong to them lies in reliable, software-based, decentralized systems. These are resilient to external influences, disruptions and also attacks. The data-based services must be transparent and interpretable so that the causes and reasons for decisions to act can be understood and traced. To achieve this, a communication network with low latency, high reliability and security must be continuously developed to provide cloud computing resources on demand. One challenge is to determine when and which computations take place at the edge of the network (edge computing) and when data is transferred to cloud computing functions. And new models of system programming are needed, especially for decentralized resource coordination, authorization, evidence, proof of compliance with relevant regulations, and accountability.

    Fields of research

    Research results

    fortiss brochure

    fortiss Brochure pdf

    Download pdf - german​​​​​​​

    Annual Report AI Engineering 2020

    Local research and innovation alliances

    Bavarian Ministry of Economic Affairs, Regional Development and Energy
    UnternehmerTUM
    German Aerospace Center (DLR)
    Fraunhofer Gesellschaft
    The Bavarian Research Institute for Digital Transformation (bidt)
    Logo IBM
    Technical University of Munich TUM
    Ludwig-Maximilians-University Munich LMU
    eit European Institute of Innovation and Technology

    Strategic research partnerships

    CGC, Aarhus University, Denmark
    EPF Lausanne, Switzerland
    Shandong University, China
    Chinese Academy of Science, China
    University of California Berkley
    IBM research
    SRI International
    Blekinge Institute of Technology, Sweden
    University Lusófona, Portugal
    Software Engineering Institute, Shanghai
    Université Grenoble Alpes VERIMAG