Geosynchronous Satellite Maneuver Classification and Orbital Pattern Anomaly Detection via Supervised Machine Learning

Abstract

Due to the nature of the geosynchronous (GEO) orbital regime, where space objects orbit the Earth once per sidereal day, GEO satellites can appear fixed to a position in the sky when observed from the Earth’s surface. This unique orbital characteristic makes GEO satellites ideal for telecommunications missions that require Earth-fixed antennas to send and receive signals, such as television broadcasts or military communications. To maintain their position relative to the Earth’s surface, GEO satellites must station-keep, or regularly expend onboard propellant to counteract the natural forces in the near-Earth space environment that perturb their orbital trajectories. Less frequently, GEO satellites perform maneuvers to alter their orbital characteristics more drastically. One such maneuver is a longitudinal shift: changing a GEO satellite’s sub-satellite point from one position on the Earth’s equator to another. Such a maneuver often requires both a series of impulsive thrusts and a period of natural drift.

This work describes an approach for detecting the components of longitudinal shift maneuvers—including the patterns associated with initiating and ending eastward and westward drifts—using convolutional neural networks trained on publicly available two-line element (TLE) data from the U.S. Space Command’s (SPACECOM) space object catalog. A method for converting TLE data to geographic position histories—longitude, latitude, and altitude positions over time in the Earth-centered, Earth-fixed geographic reference frame—and labeling longitudinal shift maneuvers by inspection is described. A preliminary maneuver detection algorithm is designed, trained, and tested on all GEO satellites in orbit from January 1 to December 31, 2020. Performance metrics are presented for algorithms trained on two different training data sets corresponding to five and ten years’ worth of geographic position time-histories labeled with longitudinal shift maneuvers.

When detected, longitudinal shift maneuvers can be used to identify anomalous behavior in GEO. In this work, a satellite’s behavior is considered nominal if it adheres to the satellite’s pattern of life (PoL)—its previous on-orbit behavior made up of sequences of both natural and non-natural behavioral modes, including routine station-keeping, other on-orbit maneuvers, and uncontrolled motion—and anomalous if it deviates from the satellite’s PoL. Identifying anomalous satellite behavior is of critical interest to space situational awareness (SSA) system operators, who may choose to task their sensors to obtain more observations of anomalous behavior, and satellite operators themselves, who may wish to diagnose its root cause. Applications of this work for international space policymaking, including the development of on-orbit norms of behavior and the distribution of spectral and physical space in GEO, is also discussed.

This work describes an approach for detecting the components of longitudinal shift maneuvers—including the patterns associated with initiating and ending eastward and westward drifts—using convolutional neural networks trained on publicly available two-line element (TLE) data from the U.S. Space Command’s (SPACECOM) space object catalog. A method for converting TLE data to geographic position histories—longitude, latitude, and altitude positions over time in the Earth-centered, Earth-fixed geographic reference frame—and labeling longitudinal shift maneuvers by inspection is described. A preliminary maneuver detection algorithm is designed, trained, and tested on all GEO satellites in orbit from January 1 to December 31, 2020. Performance metrics are presented for algorithms trained on two different training data sets corresponding to five and ten years’ worth of geographic position time-histories labeled with longitudinal shift maneuvers.

When detected, longitudinal shift maneuvers can be used to identify anomalous behavior in GEO. In this work, a satellite’s behavior is considered nominal if it adheres to the satellite’s pattern of life (PoL)—its previous on-orbit behavior made up of sequences of both natural and non-natural behavioral modes, including routine station-keeping, other on-orbit maneuvers, and uncontrolled motion—and anomalous if it deviates from the satellite’s PoL. Identifying anomalous satellite behavior is of critical interest to space situational awareness (SSA) system operators, who may choose to task their sensors to obtain more observations of anomalous behavior, and satellite operators themselves, who may wish to diagnose its root cause. Applications of this work for international space policymaking, including the development of on-orbit norms of behavior and the distribution of spectral and physical space in GEO, is also discussed.