Autonomous Detection of Satellite Orbital Anomalies Through Statistical Behavioural Analysis

The proliferation of Resident Space Objects (RSOs) within the near-Earth orbital regime presents serious challenges for the sustainability and safety of space operations. Accurate tracking and cataloguing of RSOs are critical for Space Situational Awareness (SSA) and Space Traffic Management (STM). These tasks are hindered by unknown maneuvers of active satellites, which can lead to difficulties in correlating new observations with existing catalogued data. To address this issue, the study proposes a novel algorithm to detect satellite orbital anomalies based on a statistical analysis of its operational behaviour. Specifically, the algorithm is designed to detect anomalous or non-nominal satellite activities through the comparison of its most recent Pattern of Life with its current behaviour. The integration of statistical behavioural models into SSA pipelines provides a scalable solution for maintaining catalogue integrity in the increasingly congested orbital environment, by identifying satellite orbital anomalies.