Robust estimation of shape invariants

We develop a feature-based shape- and motion-estimation scheme for moving radar targets based on the solution of object-image relations (OIRs) for incomplete and noisy data. Unlike previous OIR-based techniques for the radar moving-target-imaging problem, our approach is robust to 1) missing data, 2) so-called “slippery scatterers” whose location moves with respect to the radar viewing aspect, and 3) extreme outliers (scattering centers that are inconsistent with the assumed physical or geometric models). The new method is based on exploiting the underlying structure of the model equations, that is, far-field radar data matrices can be decomposed into a summation of low-rank and sparse matrices.