NPF算法在X射线脉冲星导航中的应用研究

针对X射线脉冲星导航中航天器模型的强非线性、高阶模型不确定性等问题，提出应用非线性预测滤波(NPF)算法实时估计航天器的轨道信息。首先，建立具有模型不确定性的X射线脉冲星导航定轨指标函数，优化得到满足指标函数最小的系统模型误差值，通过降低模型不确定性的影响来提高航天器自主定轨精度。对STK生成的“火星探路者”和“金星快车”及“北斗一号”三种航天器轨道数据进行分析，仿真结果表明，该算法比EKF算法具有更高的定轨精度，能够满足深空以及近地轨道航天器的自主定轨精度指标要求。

Study on the Application of NPF Algorithm in the X Ray Pulsar Navigation

In this paper, a Nonlinear Predictive Filter (NPF) based on the minimum model error criterion is proposed to estimate the orbit information of spacecrafts in real time, which can solve problem contained in the X-ray pulsar navigation, such as strong nonlinearity of spacecraft system model and high-order model uncertainty. Firstly, A navigation orbit index function is established based on the models with uncertainty,and the model error is calculated which minimizes the index function,thus the influence caused by model uncertainty can be reduced and eventually the accuracy of autonomous orbit determination (OD) for spacecrafts can be further improved. The orbits of Path Finder,Venus Express and Beidou-01B are adopted to validate the navigation method. The simulation results demonstrate that the proposed method based on NPF has higher precision than EKF, and it can meet the requirements of autonomous OD accuracy for different orbit types of spacecrafts.