Recent Developments in Inverse Problems - Abstract
Wagner, Roland
As the image quality of modern ground based telescopes like the planned European Extremely Large Telescope (E-ELT) is degraded by effects of atmospheric turbulences, they depend heavily on Adaptive Optics (AO) systems. Using measurements of incoming wavefronts from guide stars, AO systems reconstruct the turbulence above the telescope and derive the shape of deformable mirror(s) (DM). Most types of AO systems rely on measurements from artificially created laser guide stars (LGS) as the sky coverage with bright stars is low. Unfortunately, measurements from LGS suffer from additional effects, e.g., spot elongation, which affect the reconstruction of the atmosphere or the shape of the DM.
We present a new reconstruction method for Ground Layer Adaptive Optics (GLAO), based on the Bayes approach. In GLAO, several guide stars, each associated to a wavefront sensor, and a single mirror is used for the correction of the turbulence in the layer closest to the ground, where usually most of the atmospheric turbulence is located. Such a system uses a combination of natural and laser guide stars. As spot elongation is a well-documented effect when observing an LGS, one can model it mathematically in the reconstruction approach. In the Bayes approach it is natural to model the spot elongation as a specific noise distribution. Contrary to the standard approach, including the noise distribution of the spot elongation directly into the reconstruction of the atmosphere, e.g., by the minimization of an appropriate Tikhonov functional, and thus solving a large coupled system of linear equations, we aim for a compensation of spot elongation in a separate preprocessing step. Together with a cumulative reconstructor (CuReD), this results in a new linear and fast GLAO reconstructor.