Smart Optics for Satellite Applications

Lead Organisation: University of Oxford
Project Lead: Karen Hampson

Partners: Surrey Satellite Technology Limited (SSTL)

A significant challenge in the production of earth observation satellites is the alignment of the telescope optical components. A misaligned telescope can lead to aberrated images and is a major cause of failure of satellite missions with optical imaging payloads. The current industrial approach to aligning the telescope optics during the build phase is manual. For example, in the alignment of a Ritchey-Chrétien telescope, the secondary mirror is aligned to the primary mirror using adjustments informed by measurements from camera images of alignment laser beams and from live interferometry. These highly skilled heuristic techniques can take many days. The objective of this project was to develop an automated alignment algorithm to reduce the alignment time of a Ritchey-Chrétien telescope from many days to minutes.

The telescope prototype developed for testing of the alignment process is shown in the Figure below. The secondary mirror is attached to a high-precision hexapod and an interferometer is used to determine the misalignment of this mirror using Zernike aberration coefficients. Zemax modelling of the effect of movement of the secondary mirror on the Zernike coefficients revealed that tip, tilt, defocus and coma are the dominant aberrations affected.

Telescope construction showing the primary and secondary mirrors, flat mirror, hexapod and interferometer

Example results of the automated alignment procedure are show in the Figure below. The algorithm is written in Python. Initially tip and tilt are corrected by decentring and rotating the secondary mirror. Then defocus is corrected by translating the secondary mirror along the optical axis. Finally, coma is corrected by rotating the secondary mirror about its centre of curvature. Typically the Zernike coefficients can be reduced to single figures in under five minutes.

Example results from alignment procedure. (a) Reduction of Zernike coefficients. (b) Phase map before and after alignment

The team has developed a fully automated alignment algorithm for a Ritchey-Chrétien telescope that will reduce the in-factory alignment time from days to minutes. This will reduce the cost and time required when building the telescope. The same procedure can be applied to other telescope designs.

The work carried out during this project has enabled the University of Oxford to obtain further funding from an EPSRC Impact Acceleration Account. This will allow Oxford to develop an alignment algorithm that can be used on board a satellite mission, using live images of the earth. Surrey Satellite Technology Ltd. (SSTL) is currently developing a satellite with a telescope that deploys and aligns autonomously in-orbit. It is envisaged that the new algorithm will be used on this satellite.