DO-178C cert-kit for airborne machine learning to be researched by Intelligent ArtifactsNews
April 22, 2021
NEW YORK, N.Y. Intelligent Artifacts announced that it has been selected for a Phase I SBIR contract award by the US Air Force to conduct research and development (R&D) in pursuit of a "Blue Sky" opportunity for airborne Machine Learning (ML) that meets strict technical standards for safety in aviation.
Certifying artificial intelligence and ML systems in airborne platforms has proven to be a challenge for manufacturers. The reasons and scenarios that lead an ML-powered algorithm to make a certain decision in place of a human is difficult to certify for safety. Understanding why and how a decision was made is what makes ML certifiable.
"One of the most difficult hurdles in satisfying aviation safety standards is the ability to trace predictions back to source data," says Intelligent Artifacts CEO, Nick Cianciolo. "Our founder, Sevak Avakians, along with the rest of the team at IA, has created a new category of 'Traceable Machine Learning' which does just that."
According to the company, the overall goal of this Phase I project is to begin third-party vetting of Intelligent Artifacts' software and creating a plan to build a DO-178C Cert-kit. Development of a DO-178C certifiable solution will allow IA to integrate their software offering into full solution stacks needing to be certified by the FAA for deployment on aircraft.