J008: Fault estimation of satellite RW using covariance based adaptive unscented Kalman filter скачать в хорошем качестве

J008: Fault estimation of satellite RW using covariance based adaptive unscented Kalman filter

Rahimi A., Kumar K., Alighanbari H. (2017, May) Fault Estimation of Satellite Reaction Wheels Using Covariance Based Adaptive Unscented Kalman Filter, Acta Astronautica, from Elsevier BV, Volume 134, Issue N/A, pp. 159-169, doi: 10.1016/j.actaastro.2017.02.003 This research paper introduces the Covariance Matching Adaptive Unscented Kalman Filter (CAUKF), a novel algorithm designed to monitor the health of satellite reaction wheels. These wheels are critical components for attitude control, yet they are susceptible to hardware malfunctions that can jeopardize entire missions. The authors propose a method that automatically adjusts state covariance matrices to rapidly identify non-measurable system failures without requiring prior knowledge of performance data. By utilizing an adaptive fault annunciation metric, the filter can effectively track abrupt, transient, intermittent, and incipient faults that standard filters often miss. Extensive simulations demonstrate that this approach significantly reduces estimation errors and improves tracking speed compared to traditional models. Ultimately, the study provides a robust framework for developing fail-safe satellites where physical hardware redundancy is limited by space and power constraints.