Mesure de front d'onde post-coronographique à haute précision pour l'imagerie à haut contraste : application sol et espace.

摘要

Performing an exoplanet direct detection means being able to image an object as faint as an extra-solar planet (10^6 to 10^10 times fainter to its host star) very close to its parent star (0,1’’ to 1’’). Such levels of contrasts can be achieved using a high-contrast imaging technique such as coronagraphy, which suppress most of the star light without modifying the signal coming from the planet. In the case of ground-based observation, coronagraphy must be coupled with an extreme adaptive optic loop (XAO) to compensate for the optical aberrations, especially the ones induced by the atmospheric turbulence, that strongly decreases the coronagraph performance. After compensation of the turbulence by the XAO loop and most of the star light removed by the coronagraph, the ultimate limitation of high contrast imaging systems lies in its quasi-static aberrations that creates a residual signal which limits the achievable contrast on the scientific detector. To increase the achievable contrast on the detector, these aberrations must be compensated for, ideally using focal plane data recorded from the scientific detector to avoid differential aberrations. The aim of this thesis was to develop a focal-plane wavefront sensor (WFS) dedicated to the estimation of quasi-static aberrations in high contrast imaging systems. This WFS, called COFFEE, consist in a extension of phase diversity to coronagraphic imaging systems. It estimates the aberrations both upstream and downstream of the coronagraph using coronagraphic focal plane images that differ from a known diversity aberrations introduced upstream of the coronagraph. During this research work, COFFEE has been developed, tested using numerical simulations, and demonstrated on an in-house bench. Considering the limitations of the estimation accuracy, COFFEE’s formalism has then been modified to allow it to estimate high frequencies aberrations with nanometric precision, even with an error on the diversity phase thanks to the use of a myopic estimation. Besides, COFFEE has been extended to the estimation of amplitude aberrations. This extended version of COFFEE has been successfully used on SPHERE to optimize the contrast on the scientific detector of the instrument using COFFEE in a dedicated compensation process. Lastly, a new compensation method relying on an energy minimization approach (Dark Hole) that can be used with COFFEE has been developed in order to reach very high contrast levels on the scientific detector.