A Novel Methodology to Construct Digital Twin Models for Spacecraft Operations Using Fault and Behaviour Trees

Successful satellite data reception requires the nominal operation of the ground stations in charge of their health monitoring as much as the spacecrafts themselves. Although the concept of Model-Based Diagnosis (MBD) in the field of autonomous systems -such as satellites, has long been researched and developed, that is not the case for their ground systems. Both satellites and ground stations operate autonomously. The latter however, are not equipped with the advanced Fault Detection, Isolation and Recovery (FDIR) capabilities one finds today on-board all orbiting spacecrafts. The aim of the study presented in this paper is the improvement of ground stations' operational diagnostics by providing the operators with ad-hock, Operations-Dedicated Models (ODMs). The latter serve as a basis for the construction of the system's Digital Twin (DT) models. These models allow the operators react more quickly and more precisely to alarms raised by the station. By helping the operators identify the malfunction and correct it in the quickest delays, they can avoid loosing the next satellite telemetry (TM) data, thus saving precious time and costs. This would increase both the availability and maintainability of the system. In a larger framework, ODMs are ideally concurrently built and connected with the engineering and safety models of the system, in a sort of virtual continuous improvement loop. While the utter purpose of ODMs is their usage as the system's DTs during operations, they also contribute to the stations' architecture and robustness continuous improvement, through increasing its fault detection and mitigation capabilities.