Atmospheric Density Uncertainty Quantification for Satellite Conjunction Assessment

TL;DR

This paper introduces a novel method for quantifying atmospheric density uncertainties and incorporating them into satellite conjunction assessments, improving collision risk estimates by accounting for density error correlations.

Contribution

It presents a dynamic reduced-order density model that efficiently estimates density uncertainties and propagates them for collision probability calculations, including cross-correlation effects.

Findings

Density uncertainties significantly impact collision probability estimates.

Ignoring cross-correlation can lead to severe underestimation of collision risk.

The method improves the accuracy of conjunction assessments by incorporating density error dependencies.

Abstract

Conjunction assessment requires knowledge of the uncertainty in the predicted orbit. Errors in the atmospheric density are a major source of error in the prediction of low Earth orbits. Therefore, accurate estimation of the density and quantification of the uncertainty in the density is required. Most atmospheric density models, however, do not provide an estimate of the uncertainty in the density. In this work, we present a new approach to quantify uncertainties in the density and to include these for calculating the probability of collision Pc. For this, we employ a recently developed dynamic reduced-order density model that enables efficient prediction of the thermospheric density. First, the model is used to obtain accurate estimates of the density and of the uncertainty in the estimates. Second, the density uncertainties are propagated forward simultaneously with orbit propagation to include the density uncertainties for Pc calculation. For this, we account for the effect of cross-correlation in position uncertainties due to density errors on the Pc. Finally, the effect of density uncertainties and cross-correlation on the Pc is assessed. The presented approach provides the distinctive capability to quantify the uncertainty in atmospheric density and to include this uncertainty for conjunction assessment while taking into account the dependence of the density errors on location and time. In addition, the results show that it is important to consider the effect of cross-correlation on the Pc, because ignoring this effect can result in severe underestimation of the collision probability.