Robust Model-based Fault Diagnosis of Thruster Faults in Spacecraft

Abstract

The research work presented in this paper addresses the design and validation of a robust model-based fault diagnosis system to detect and isolate any thruster fault occurring in the thruster-based propulsion system of the chaser spacecraft during rendezvous on a circular orbit around the Mars planet. The REFERENCE mission is the ESA Mars Sample Return (MSR) mission. Key features of the proposed method are the use of a Hi/H∞ filter for robust fault detection and a bank of unknown input observers jointly used with a dot product of vectors strategy for fault isolation. Robust performance over a large range of uncertainties (a total of 36 uncertainties is considered) is formally demonstrated through the μ3 formalism.