Institute of Embedded Systems

Modern systems and avionics in aircraft for safety-critical functions are already exposed to extremely high requirements today. This affects design, run-time behavior and continuous testing for qualification.

Current avionics architectures must evolve from single-core towards multi- and many-core processors due to obsolescence. This requires new ways of data exchange with significant impact on the distribution of safety-critical functions (such as flight control systems) on the computing platform, run-time behavior and synchronised communication between function blocks. Only very few processors are targeting that market but are black-box. The avionics platform design needs radical innovation. Separation of safety-critical from other functions usually leads to physical separation on the platform. For many-core architectures, this separation must be considered on processor/electronics level. Additionally, the number of applications is rising fast. Hence, new methods and tools to design, model, develop and validate such types of systems are required. Such design, development and qualification aspects will become relevant also in other domains such as autonomous driving vehicles.

New design and analysis methods for safety-critical systems and software on many-core architectures shall be developed, taking modern flight control system platforms as use-case. These methods shall especially consider runtime and qualification aspects.

The Institute of Embedded Systems delivers into the project the requirements and capabilities of many-core architectures through the layer of implementation (bottom-up approach). The Institute of Aircraft Systems Engineering provides a realistic, complex reference system as well as system and architecture requirements (top-down approach). A corporate research platform will be developed in the form of a virtual research avionics platform that will run parallelised, safety-critical flight control system applications based on many-core technology. New methods and tools will allow a systematic, model-based system design and simulation-based validation.