Despite the advancement of machine learning and artificial intelligence, safety still constitutes the main hurdle for supporting high levels of autonomy in domains such as self-driving cars, where more than 400* car accidents** involving autonomous functionalities were reported last year*. The design process according to safety standards typically assumes a correct and complete system specification. For autonomous systems, it is often impossible to show that the specification is complete, e.g. due to the underspecified environment and evolving behavior. This extremely challenges current safety engineering practices to reason about risk and uncertainty at operation time for decision making.

Sponsor