Design and implementation of a low cost attitude determination and control system with novel star tracker algorithm.

The attitude determination and control system determines and controls the orientation of the spacecraft. This system is crucial in the majority of space missions to e.g. point a camera to a star or direct an antenna to a ground station. Increasingly complex missions drive the need for higher accuracy, while the growing number of small spacecraft requires high robustness and low computational cost. This work focusses on the star tracker, a sensor that takes an image of the stars and compares it to a database with known star positions to determine the spacecraft attitude. Algorithms are developed with high accuracy, high robustness and low computational cost. A full attitude determination and control system for a class of nanosatellites, called CubeSats, is also presented.

The centroiding algorithm determines the centroid of stars in the camera image. The developed algorithm uses closed form expressions to fit a model to the measured star data. The use of model fitting leads to high accuracy, while the closed form expressions keep the computational cost low. The accuracy is in the range of the most accurate algorithms and the computational cost is in the range of the fastest algorithms.

The lost-in-space algorithm matches the stars in the camera image to stars in the database. The developed algorithm, which depends on the Shortest Distance transform is very robust to false stars, distortions in the image and missing stars. On top of that, it offers a reliable quality value for the result, which further increases the robustness.

The tracking algorithm finds the transformation values between camera and database stars and determines the attitude based on these values. The AIM algorithm developed in this work is the fastest tracking algorithm available. Its novel approach to solving the tracking problem can be exploited to increase the robustness and decrease the computational cost. 

The algorithms deliver similar accuracy as the most accurate state of the art algorithms. Their computational cost is lower and they have higher robustness. These algorithms and the reaction wheels designed at KU Leuven are enablers for a high accuracy attitude determination and control system for CubeSats.

A low-cost high accuracy attitude determination and control system for CubeSats is under development at KU Leuven. The system delivers high pointing accuracy for low volume, weight and power consumption. It opens up potential for small satellite missions that have higher demands on the pointing accuracy. The system is introduced and its performance is analysed in simulations.