ILAS2016 — 11–15 July 2016 — KU Leuven, Belgium

20th Conference of the International Linear Algebra Society (ILAS)

20th ILAS Conference

In minisymposium: Image Restoration and Reconstruction

Fri 11:30–12:00, Room AV 01.12
Fast iterative algorithms for wide field of view Adaptive Optics
Daniela Saxenhuber (Johannes Kepler University Linz)
Joint work with Ronny Ramlau (Johannes Kepler University Linz)

Adaptive Optics (AO) systems in large ground-based telescopes aim at mechanically correcting for atmospheric distortions. The incoming light is reflected on a deformable mirror in order to improve the image quality before the wavefront propagates to the scientific camera. The deformable mirror has to be controlled in real time, i.e. at 500 - 1000 Hertz. In wide field of view AO, light from several astronomical objects or artificially created laser beacons is measured in order to determine the optimal shape of the deformable mirror. This problem is referred to as atmospheric tomography and is an ill-posed inverse problem. For the new generation of Extremely Large Telescopes (ELT) with mirror diameters of up to 40 m, such complex AO systems demand reconstruction times of around 1 ms for several thousand unknowns. In this talk, we present fast iterative reconstruction algorithms for the solution of this limited-angle tomography problem. Furthermore, we include real-life effects such as tip/tilt indetermination, cone effect and spot elongation in the modelling. In addition, we present novel approaches to determine a sparse reconstruction profile for atmospheric tomography in order to reduce the computational cost even further. Simulation results are presented within an ELT-setting on OCTOPUS, the end-to-end simulation tool of the European Southern Observatory.