Infrared imaging for simultaneous Earth horizon and Sun detection provides a compact solution for spacecraft attitude determination, particularly for small satellites with strict resource constraints. This thesis investigates the use of low-resolution infrared cameras for three-axis attitude estimation in low Earth orbit, with application to the Delfi-Twin satellite platform.

A simulation framework was developed to generate realistic synthetic horizon images incorporating orbital geometry and sensor characteristics. Classical image-processing methods were implemented to extract horizon and Sun vectors and reconstruct attitude from geometric observations. Performance was assessed through extensive simulations and Monte Carlo analyses to quantify accuracy and robustness.

Results show that accurate horizon and Sun vector measurements can be obtained from low-resolution imagery, enabling stable attitude estimation over a wide range of viewing geometries. Monte Carlo simulations indicate robust performance under realistic sensor noise and partial Earth visibility conditions. These findings demonstrate that reliable attitude estimates can be achieved using only low-resolution infrared imaging combined with gyros.