基于经验加速度的低轨卫星轨道预报新方法

A New Method of Orbit Prediction for LEO Satellites Using Empirical Accelerations

研究将定轨过程中的经验加速度应用于地球低轨卫星轨道预报的新方法.利用GPS伪距观测数据和简化动力学最小二乘批处理方法对地球低轨卫星定轨,其中卫星位置、速度及大气阻力系数和辐射光压系数可以直接用于轨道预报.作为简化动力学最重要特征的经验加速度呈现准周期、余弦曲线特点,可通过傅里叶级数拟合建模.确定性动力学模型与补偿大气阻力模型误差的切向经验加速度级数拟合模型组成增强型动力学模型用于提高轨道预报精度.应用GRACR-A星载GPS伪距观测数据和IGS超快星历定轨并进行轨道预报,结果表明轨道预报初值位置精度达到0.2 m,速度精度达到1×10^(-4)m·s^(-1),预报3天位置精度优于60 m,比只利用确定性动力学模型进行预报精度平均提高2.3倍.先定轨后预报的模式可用在星上自主精确导航系统中.

A new method of orbit prediction for LEO satellites is proposed by ttsing empirical accelerations to compensate for mismodelling of deterministic force models.The position,velocity and dynamical parameters such as C_D,C_R,and empirical accelerations are calculated in the orbit determination using reduced dynamic batch Least-Squares technique and GPS pseudorange observations.Therefore the position,velocity,C_D,and C_R.can be used for prediction directly.Furthermore,the empirical accelerations are showed quasi-periodic and cosine characteristics with respect to time variable,in terms of which Fourier series were used to interpolate empirical accelerations.In this way,the fitted tangential empirical accelerations curve,which is added into deterministic force models to compensate for mismodelling of atmospheric drag model,forms enhanced-accuracy dynamic models that are used for orbit prediction.The GRACE-A real flight GPS pseudorange data and IGS Ultra-rapid products have been used to orbit determination and then the proposed method has been used to orbit prediction.The orbit determination results show that initial positioning accuracy is about 0.2 m and velocity accuracy is about 1.0 × 10^(-4)m·s^(-1).For a 72-hour orbit pass,the prediction accuracy is better than 60 m,which is averagely improved about 2.3 times when compared to conventional dynamic models without considering empirical acceleration series fitting models.The proposed orbit determination and prediction scheme are beneficial to establish advanced,even onboard,satellite autonomous navigation system.