Observation and Estimation for Space Applications: Software Algorithms and Hardware Design (Metingen en schattingen in ruimte-onderzoek: software algoritmes en mechanisch ontwerp)

The oscillations in the photometric signal of stars is determined by tiny periodic changes of their internal structure. Comparing high-precision observations of the oscillation frequencies with those predicted by theoretical stellar models to gain insight in stellar structure and evolution, is the key objective of asteroseismology. In this dissertation, some technical aspects related to the acquisition of high-precision measurements are investigated. The Mercator Advanced Imager for Asteroseismology (MAIA) is a ground-based instrument specifically designed to measure fast oscillations of compact pulsating stars in three distinct colour bands. The mechanical structure of MAIA is designed such that the optics are quasi insensitive to the temperature variations of the instrument. This dissertation further describes the thermodynamic coolers that are used to cool the detectors, including the technical challenges encountered.The PLATO space mission is proposed to ESA for a launch in 2024. The science objectives of this mission are to detect exoplanets and to characterize their host stars with asteroseismological techniques, by providing the photometric signal in one broadband for the majority of targets, and in two distinct colour bands for the few thousand brightest targets in the field of view. In order to separate stellar variability from instrumental effects, stringent requirements are set on the pointing accuracy of the spacecraft.In this dissertation, an analytical method is developed to predict the statistics of spacecraft pointing errors. Applying this method to the PLATO space mission has led to a number of recommendations concerning the selection and calibration of the onboard attitude sensors, the thermal stability of the platform and the attitude determination and control software.Due to the low disturbances encountered in space, the pointing error is dominated by sensor noise for most spacecraft. In this dissertation, a moving horizon estimator is developed to reduce the effect of sensor noise. The estimated attitude is computed by solving a least-squares problem using a horizon of measurements and a mathematical model of the spacecraft. Both the attitude and sensor calibration parameters can be estimated together, resulting in a further increase in accuracy. Our new method offers advantages compared to currently used methods in spacecraft applications.status: publishe