Optimization of Hybrid Method for Uncertainty Propagation of Non-Keplerian Motion

摘要

One of the most important technical discussions in space situational awareness is how to capture the dynamical nonlinearity to propagate uncertainty accurately. Our previous research considered the dynamical realism for accurate uncertainty propagation and proposed a hybrid method that combined advantages of the semi-analytic solutions and nonlinear state mapping tensor, i.e., state transition tensor. The main goal of this research is to investigate an optimal combination of these two methods, making a hybrid method. The goal is carried out by introducing a cost function including two key factors: computational efficiency and accuracy of uncertainty propagation. In this research, we establish eight combinations, based on second- and third-order semi-analytic solutions and the form of the tensors up to fourth order, and compare the performance of each case with the calculated costs. Through this research, we show that the combination of the third-order semi-analytic solutions and the second-order state transition tensor has the best performance in the most cases.