Methods to Detect Impact-Induced Orbit Perturbations Using Spacecraft Navigation Data

Abstract

Debris strikes on operational spacecraft are becoming more common due to increasing numbers of space objects. Sample return missions indicate hundreds of minor strikes, but rigorous analysis is often only performed when a strike causes an anomaly in spacecraft performance. Developing techniques to identify and assess minor strikes that do not immediately cause anomalous behavior can help to validate models for debris populations and aid in the attribution of future anomalies. This study develops methods to detect subtle abrupt orbit perturbations indicative of minor debris strikes. An extended Kalman filter with dynamic model compensation is used to estimate a spacecraft’s orbit state based on simulated full-state (i.e., GPS) measurements. The filter is applied to the data forward and backward in time, and then a modified Fraser–Potter smoother is used to produce a fused state estimate. Various test statistics are developed and compared to identify abrupt unexpected changes in spacecraft velocity; techniques include McReynold’s filter-smoother consistency test and the Mahalanobis distance between forward and backward filter states. A trade study is performed to investigate the performance of test statistics as a function of filter parameters, and a Monte Carlo analysis illustrates the filter’s ability to detect and estimate strikes.