From fundamental research to marketable prototypes
fortiss has established itself at the forefront of global research on key topics in
and is a recognised partner for challenging issues in software development and artificial intelligence (AI).
The institute focuses on methods and tools for developing and operating powerful software with reliable functionality, performance, resilience, persistence, security and maintainability. The focus is on integrating model-driven software development with data-driven AI programming to enable the controlled development of a new generation of increasingly autonomous and decentralised software systems.
Until now, software and systems engineering has focused on ensuring the reliability and security of relatively small, centralised and automated systems that operate in predictable environments. However, current and future requirements are increasingly shifting towards ensuring the trustworthiness of larger, dynamically networked, self-learning and constantly evolving, often autonomously acting systems. In the Software & Systems Engineering research focus area, fortiss is expanding the classic methods of model-based software and system development and integrating them with new development methods for data-driven applications.
The current focus topics in this research focus area include the structured development of trustworthy autonomous systems and the analysis and optimisation of software and system architectures. Furthermore, fortiss deals with software engineering for data-driven applications, the use of AI methods in software engineering and the validation and certification of large software systems.
The AI Engineering research area focuses on researching and developing a new generation of robust, trustworthy AI technologies that can make timely, safe decisions in uncertain and unpredictable environments. Our goal is to increase the trustworthiness and explainability of AI systems and ensure they are resilient to erroneous inputs and targeted attacks. The focus here is on processing large volumes of data as well as gaining knowledge from small amounts of data without jeopardising confidentiality and privacy.
A central concern of fortiss is the development of efficient AI systems that can be used in mission-critical and safety-relevant applications. Through a deep understanding of the underlying concepts and application scenarios and an engineering approach, we ensure that these systems offer comprehensible and transparent decision-making processes. With a focus on human-machine interaction, we develop data-based, intelligent user interfaces that meet user requirements, promote the integration and acceptance of intelligent systems and make interaction natural, intuitive and safe.
In addition, fortiss uses machine learning to research new approaches to reinforcement learning and representation learning to improve the ability of data-driven systems to learn and adapt to changing conditions. Our activities in neuromorphic computing focus on developing energy-efficient, low-latency neural networks that can be used in robotics and industrial applications in particular.
In the IoT Engineering research focus area, fortiss concentrates on the development of flexible, software-based infrastructures that can be adapted dynamically and in line with changing requirements. This requires a deep and comprehensive integration of sensor, computer and communication capabilities into existing, often heterogeneous systems as well as the software-supported penetration of traditionally physical, critical infrastructures. The basis for these intelligent infrastructures is formed by reliable, software-based, decentralised systems that are resilient to external influences, disruptions and attacks. Transparent, data- and knowledge-based services are essential to ensure authorisation, make decision-making processes traceable, coordinate decentralised resources and enable verification mechanisms for compliance with relevant regulations and accountability obligations.
The continuous development of communication networks with low latency, high reliability and security as well as the needs-based provision of cloud computing resources are central tasks of IIoT engineering. One of the major challenges is deciding when and what data is processed at the edge of the network (edge computing) and when it is forwarded to cloud services. fortiss is therefore focussing on the development of these robust, trustworthy platforms and infrastructures to enable a connected, data-based future. By combining edge and cloud computing, we create the basis for efficient, secure and scalable IoT solutions that meet the requirements of modern, dynamic environments.
