Parameter estimation of a gyroless micro-satellite from telemetry data

Abstract

This paper presents a new approach for the simultaneous inertia and actuator alignment parameter estimation. One of the strengths of the proposed estimation algorithm consists in its ability to work directly with the star tracker attitude measurements and it is therefore also suitable for small and medium size satellites, where the gyroscope measures are not reliable or not available. In the algorithm, the satellite inverse dynamics model is exploited for the estimation cost function to take advantage of the linearity with respect to the main satellite parameters. The overall proposed solution is based on the instrumental variable method which has been deeply adapted to take into account several practical issues. The main components of the disturbance torques are modeled and simultaneously estimated with the satellite parameters to compensate for their effect. Inspired by the prefiltering technique used in the optimal refined instrumental variable method, a tailor-made filter design is proposed and implemented to deal with noisy measurements and the unmodeled disturbance torques. Such prefilter considerably improves the algorithm performance by achieving low-variance parameter estimates, while still requiring little to no tuning. A final but important novelty of this work consists in proposing an effective solution for estimating and compensating synchronism errors in the sensor measurements. This latter aspect has never been considered in the literature, although the lack of synchronism has a non-negligible effect on the accuracy of the parameter estimates. The proposed method is analyzed and validated via Monte Carlo simulations using data coming from a high-fidelity satellite simulator designed by the French space agency, and the results show the effectiveness of the algorithm.