Statistical Decomposition and Machine Learning to Clean In Situ Spaceflight Magnetic Field Measurements

• DOI: 10.1029/2023gl103626
• Publication Date: 2023-06-30T05:22:00Z
• Authors: M. G. Finley; T. A. Bowen; M. Pulupa; A. Koval; D. M. Miles
• Keywords: Reaction wheel

Abstract Robust in situ magnetic field measurements are critical to understanding the various mechanisms that couple mass, momentum, and energy throughout our solar system. However, spacecraft on which magnetometers often deployed contaminate via onboard subsystems including reaction wheels magnetorquers. Two can be at different distances from determine an approximation of interfering for subsequent removal, but constant data streams both impractical due power telemetry limitations. Here we propose a method identify remove time‐varying interference sources such as using statistical decomposition convolutional neural networks, providing high‐fidelity even cases where dual‐sensor not constantly available. For example, measurement interval Parker Solar Probe outboard magnetometer experienced 95.1% reduction wheel following application proposed technique.