Iterative parallel registration of strongly misaligned wavefront segments

The paper presents an algorithm for the precise registration of multiple wavefront segments containing large misalignment and phase differences. The measurement of a wavefront with huge dynamics or a large aperture size can be carried out in multiple Shack-Hartmann sensor measurements of segments of the wavefront. The registration algorithm is flexible with respect to the shape of the wavefront and can reconstruct a plane as well as divergent wavefronts, making it suitable for freeform wavefronts. The algorithm enables parallel registration of the wavefront segments which is carried out in an iterative manner to compensate for large misalignment errors. A simulative analysis of the proposed algorithm compares its performance to a fast parallel registration (FPR) algorithm and the established iterative closest point (ICP) algorithm. For a sensor misalignment of up to 100 μm and 3 mrad the algorithm registers a plane and a divergent wavefront with a precision that is a factor 4 and 12 better than the registration precision of the FPR and ICP algorithm.