Session
Weekend Session 1: Advanced Concepts - Research & Academia I
Location
Utah State University, Logan, UT
Abstract
To continue the growing success of scientific discovery through deep space exploration, missions need to be able to be able to travel further and operate more efficiently than ever before. To ensure resilience in this capability, on-board autonomous fault mitigation methods must be developed and matured. To this end, we present a system for cross-subsystem fault diagnosis of satellites using spacecraft telemetry. Our system leverages a combination of Kalman Filters, Autoencoders, and Causality algorithms. We test our system for accuracy against three data sets of varying complexity levels, along with baseline testing. Additionally, we perform an ablation study to evaluate on-board tractability.
Autonomous System-Level Fault Diagnosis in Satellites Using Housekeeping Telemetry
Utah State University, Logan, UT
To continue the growing success of scientific discovery through deep space exploration, missions need to be able to be able to travel further and operate more efficiently than ever before. To ensure resilience in this capability, on-board autonomous fault mitigation methods must be developed and matured. To this end, we present a system for cross-subsystem fault diagnosis of satellites using spacecraft telemetry. Our system leverages a combination of Kalman Filters, Autoencoders, and Causality algorithms. We test our system for accuracy against three data sets of varying complexity levels, along with baseline testing. Additionally, we perform an ablation study to evaluate on-board tractability.
