Correction Of Misalignment Dependent Aberrations Of The Hubble Space Telescope Via Phase Retrieval

The optics of the Hubble Space Telescope consist essentially of a primal), and a secondary mirror. Measurements from three wavefront sensors are telemetered to the ground where a series of computations are made to determine whether the telescope is performing within specification. If not, optimum secondary mirror realignment parameters are computed based on interferometrically determined aberrations present in the wavefront. This method, which is the primary method for wavefront correction of the telescope requires at least two of the three wavefront sensors to be available. To provide additional reliability a backup scheme has been implemented which makes use of a phase retrieval algorithm to compute secondary mirror misalignments from measured point spread function data at two different field locations. The point spread function data is obtained from two scientific instruments, the Faint Object Camera and the Wide Field Planetary Camera. After image centroiding to remove unnecessary tilt, and pixel deconvolution to remove detector effects, comparison pupil functions are written in terms of annular Zernike Polynomials. At the two field locations of the instruments, artificial point spread functions are generated from Fast Fourier Ransforms (FFT) of the respective pupil functions and the mean square differences between the actual PSF's and constructed PSF's are minimized with respect to the Zernike Polynomial coefficients. Thus two sets of coefficients are obtained one for each of the field positions. Statistical regression analysis is then used to determine the misalignments as a function of the aberration coefficients.