Spacecraft Attitude Determination with Sun Sensors, Horizon Sensors and Gyros: Comparison of Steady-State Kalman Filter and Extended Kalman Filter

Abstract

Attitude determination, along with attitude control, is critical to functioning of every space mission. In this paper, we investigate and compare, through simulation, the application of two autonomous sequential attitude estimation algorithms, adopted from the literature, for attitude determination using attitude sensors (sun sensor and horizon sensors) and rate-integrating gyros. The two algorithms are: the direction cosine matrix (DCM) based steady-state Kalman Filter, and the classic quaternion-based Extended Kalman Filter. To make the analysis realistic, as well as to improve the attitude determination accuracies, detailed sensor measurement models are developed. Modifications in the attitude determination algorithms for estimation of additional states to account for sensor biases and misalignments are presented. A modular six degree-of-freedom closed-loop simulation, developed in house, is used to observe and compare the performances of the attitude determination algorithms.