Innovative AI Inversion Methods for Ground Penetrating Radar Applications

This work introduces an innovative Artificial Intelligence (AI)-driven microwave imaging technique for ground penetrating radar (GPR) applications. The complexity of the non-linear/ill-posed subsurface inverse scattering problem (ISP) is addressed within the System-by-Design (SbD) framework by means of suitably chosen and interconnected functional blocks enabling both a smart definition of the solution descriptors and the interactive integration of a global optimizer with a machine learning (ML)-based surrogate model. An illustrative result is shown to assess the effectiveness of the proposed method.