Abstract
This research programme has studied the possibilities and difficulties of using 50 kg microsatellites to perform remote imaging of the Earth. The design constraints of these missions are quite different to those encountered in larger, conventional spacecraft. While the main attractions of microsatellites are low cost and fast response times, they present the following key limitations: Payload mass under 5 kg, Continuous payload power under 5 Watts, peak power up to 15 Watts, Narrow communications bandwidths (9.6 / 38.4 kbps), Attitude control to within 5°, No moving mechanics. The most significant factor is the limited attitude stability. Without sub-degree attitude control, conventional scanning imaging systems cannot preserve scene geometry, and are therefore poorly suited to current microsatellite capabilities. The foremost conclusion of this thesis is that electronic cameras, which capture entire scenes in a single operation, must be used to overcome the effects of the satellite's motion. The potential applications of electronic cameras, including microsatellite remote sensing, have erupted with the recent availability of high sensitivity field-array CCD (charge-coupled device) image sensors. The research programme has established suitable techniques and architectures necessary for CCD sensors, cameras and entire imaging systems to fulfil scientific/commercial remote sensing despite the difficult conditions on microsatellites. The author has refined these theories by designing, building and exploiting in-orbit five generations of electronic cameras. The major objective of meteorological scale imaging was conclusively demonstrated by the Earth imaging camera flown on the UoSAT-5 spacecraft in 1991. Improved cameras have since been carried by the KITSAT-1 (1992) and PoSAT-1 (1993) microsatellites. PoSAT-1 also flies a medium resolution camera (200 metres) which (despite complete success) has highlighted certain limitations of microsatellites for high resolution remote sensing. A reworked, and extensively modularised, design has been developed for the four camera systems deployed on the FASat-Alfa mission (1995). Based on the success of these missions, this thesis presents many recommendations for the design of microsatellite imaging systems. The novelty of this research programme has been the principle of designing practical camera systems to fit on an existing, highly restrictive, satellite platform, rather than conceiving a fictitious small satellite to support a high performance scanning imager. This pragmatic approach has resulted in the first incontestable demonstrations of the feasibility of remote sensing of the Earth from inexpensive microsatellites.