ORBIT PREDICTION ACCURACY USING VASICEK, GAUSSMARKOV, AND RANDOM WALK STOCHASTIC MODELS

摘要

Accurate predicted orbits are a significant concern for space situational awarenessand catalog maintenance. A key element of this is estimation of the ballisticcoefficient when its behavior is dynamic or unknown. This paper analyzesthe use of the Vasicek, Gauss-Markov, and random walk stochastic models tosolve for the ballistic coefficient during orbit determination. Different ballisticcoefficient profiles are simulated and the effect on the definitive and predictedorbit accuracy is compared to truth under ideal tracking conditions and a typicalcatalog maintenance tracking schedule. Guidelines for applying these modelsto operational orbit determination problems are then proposed.