Multi-Tier Labeling and Physics-Informed Learning for Orbital Anomaly Detection at Scale

TL;DR

This paper introduces a multi-tier labeling approach combining physics rules, Kalman filtering, and calibration to generate large-scale orbital anomaly labels, enabling effective training of a high-recall transformer model for satellite anomaly detection.

Contribution

It presents a novel multi-source labeling cascade and a transformer model that significantly improves anomaly detection recall in LEO satellite data.

Findings

IMM-UKF detects 42.6x more anomalies than rule-based methods.

Transformer achieves 55.4% maneuver recall and 62.8% decay recall.

Time-delta feature improves decay recall by 107%.

Abstract

Detecting orbital anomalies, such as maneuvers, atmospheric decay, and attitude upsets, across the rapidly growing population of low-Earth-orbit (LEO) satellites is a prerequisite for collision avoidance, decay forecasting, and conjunction screening. The bottleneck is not modeling capacity but labels: there is no public ground-truth corpus of orbital anomalies, manual review does not scale to approximately 10^4 active satellites, and pure rule-based detectors trade recall for precision so aggressively that they are blind to most behavioral anomalies. We present a multi-tier labeling cascade that composes three weak supervision sources of increasing fidelity: a fast physics rule set (rule_v1), an Interacting Multiple Model Unscented Kalman Filter (IMM-UKF) bank, and a supplemental-element calibration step (supGP), to produce labels at a scale unavailable from any single source. Applied to 232M Two-Line Element (TLE) records spanning 60 years, the cascade yields 8.6M labeled sequences of length 50 (430M timesteps) over 11 features that include explicit time encoding and full mean-element state. On overlapping satellites, IMM-UKF surfaces 42.6x more anomalies than rule_v1 alone. We train a 6.5M-parameter Transformer in two stages, achieving a maneuver recall of 55.4% and decay recall of 62.8% on a held-out test set. An ablation on the time-delta feature alone yields a 107% relative improvement in decay recall. We frame the resulting model as a high-recall triage classifier whose role is to surface candidate events for downstream filtering, not to issue final attributions, and discuss the path toward a Neural-ODE-based orbital world model.