Using Digital Twins to Tame the Testing of AI

Abstract

Program test managers and test engineers should carefully consider Digital Twinning approaches for addressing training and testing challenges for Artificial Intelligence/Machine Learning (AI/ML) systems. A hybrid Hardware in the Loop (HITL) and Digital Twin (DT) architecture is discussed for a notional Cognitive EW system. This architecture may provide effective training and testing for complex AI/ML systems that incorporate extensive Cyber-Physical interactions. Considerations for generating realistic RF test environments for Cognitive EW systems are also considered.