Advanced Methods of Certification

Advanced Methods of Certification (AMC)

Enabling the safe usage of AI in aerospace systems

Open Positions

If you are interested in collaborating, please contact  Prof. Daw with a topic proposal and your CV.

Vision

Flying has been one of the safest forms of transportation for decades. This important achievement can be attributed to certification processes that push for high quality standards throughout the development, testing, operation, maintenance and inspection of aircraft. Currently, we are at an inflection point where new technologies such as AI could revolutionize the aerospace industry. Unmanned aerial vehicles are already a reality.  Autonomous delivery drones, flying cabs and similar applications could soon follow. However, due to safety concerns, such technologies have not yet been adopted in the aerospace industry, and rightly so. However, there are many efforts by the FAA, EASA, NASA and other government agencies to create certification standards to ensure that these technologies are flight-ready.

In the AMC Lab, we work on engineering methods to develop AI-enabled systems that are safe and reliable enough to be used in a safety-critical application. In other words, certifiable. Our work is based on three research areas: The first concerns shifting the system specification paradigm from function-driven to data- and learning-property-driven for data-intensive applications. The second area is concerned with ensuring an AI-enabled system from data selection through training, implementation, testing, operation, and evolution. In addition, this area includes building continuous quality assurance through the development of failure mitigation strategies. The third area involves the development of engineering solutions based on the results of the other research areas to enable the approval of AI-enabled systems.

Projektmitarbeiter und -mitarbeiterinnen

To the top of the page