Improving the performance of adaptive optics systems with optimized control methods

摘要

This thesis investigates control aspects of adaptive optics (AO), a technology to compensate the rapidly changing distortions that affect light after propagating through the turbulent atmosphere. In particular, two different astronomical applications are considered: partial correction of wide fields (needed for surveys) and high accuracy correction of very small fields (needed for detecting faint companions, like exoplanets). The performance of typical current and future AO systems has been analyzed through numerical simulations, and methods to improve their performance have been studied. In the first part of the thesis, an optimum compensation of wide fields has been shown to be achievable by traditional control methods. The latter part of the thesis concentrates on the nonlinearity issues of a pyramid wavefront sensor (P-WFS) shown in earlier works to be a promising choice for the accurate small field compensation (extreme adaptive optics) due to its better sensitivity for low frequency wavefront distortions. Two novel methods to deal with the P-WFS nonlinearity effects are presented in this thesis. The first is a theoretical model based and computationally intensive method based on directly inverting the P-WFS signal model. The second method is a heuristic, computationally efficient method combining the a priori information of the atmosphere, the P-WFS signal model and experimentally obtained interaction matrices describing the system behavior. It is shown in simulations that the latter method, based on compensating the P-WFS loss of sensitivity, dramatically improves the system performance (compared to the conventional AO system control and wavefront reconstruction) in conditions where the measured wavefront aberrations are large (bad seeing and short sensing wavelengths).