Neural Networks and PCA coefficients to identify and correct aberrations in Adaptive Optics

Static and quasi-static aberrations represent a great limit for high contrast imaging in large telescopes. Among them the most important ones are all the aberrations not corrected by Adaptive Optics (AO) system, called Non-Common Path Aberrations (NCPA).An estimate of the NCPA can be obtained by a trial-and-error approach or by more sophisticated techniques of focal plane wavefront sensing.In all cases, a fast procedure is desirable to limit the telescope downtime and to repeat, if needed, the correction procedure to cope with the temporal variation of the NCPA.In this presentation, through simulated images, I will describe the application of a supervised NN for the mitigation of NCPA in high contrast imaging at visible wavelengths. I will use a dataset of simulated images on which I will perform a Principal Component Analysis (PCA) and use the "Principal Components scores" as input of a Multilayer Perceptron Neural Network (MLP NN).In this presentation I show some results of the method on different simulated conditions and preliminary results obtained on an optical bench, called GHOST, at the ESO headquarters in Garching.