Model-based Systems Engineering

Cyber-physical systems are characterized by complex and highly-interactive requirements, functions and subsystems. The interaction of these products, which are largely software-defined, with other systems as well as with our physical world, such as industrial automation, automobile, aerospace and rail, leads to a high level of complexity and places high demands on their safety and performance. Since the implementation of new functions or the correction of errors requires continuous and quickly available software updates, a system development approach that is as flexible and agile as possible is needed, but this currently poses enormous challenges in terms of securing the systems.

With this in mind, the Model-based Systems Engineering field of competence concentrates it research activities on the adaptivity, resilience and reliability of cyber-physical systems, which are designed to operate even in case of outages and manageable in terms of the complexity. To this end, we are investigating system architectures for future systems and are relying on integrated system models in order to make both the complexity manageable through the automation of development tasks and to enable the required flexibility. The focus here is on procedures for bringing forward automatic design space exploration for quality assurance in early development phases. We also offer systematic approaches for managing product variability and reusing development artefacts.

With these activities we concentrate on the development of practicable and industrially applicable solutions, combined with appropriate tool support. This enables manufacturers and suppliers to increase both their productivity and the quality of their products. We illustrate and evaluate our results using autonomous driving use cases in the fortissimo Rover demonstrator at the fortiss labs, which also serves as a platform for teaching and training opportunities.