多站仅测速实时定轨的容积卡尔曼滤波方法

Cubature Kalman filter for real-time orbit determination by multi-station with velocity data only

针对应急测控条件下低轨卫星轨道状态的实时确定问题,提出一种基于多站测速值的容积卡尔曼滤波方法。该方法采用三阶球面-径向容积准则对非线性高斯加权积分进行近似,具有比无迹卡尔曼滤波更高的滤波估计精度和数值计算稳定性。在低轨卫星轨道动力学模型中考虑了J2项非球形引力摄动和大气阻力摄动,并采用四阶龙格库塔法进行离散化处理,有效的提高了状态模型精度,并与轨道状态与测速值间的非线性量测方程一起构成了滤波系统方程。仿真实验结果表明,三站实时定位均方误差为7.059 m,定速精度为0.065 m/s,可以满足对低轨卫星实时定轨的精度需求,从而验证了本文方法的有效性。

A cubature Kalman filter algorithm for real-time orbit determination by multi-station with velocity data was proposed to get low earth orbit satellite’s(LEO)real-time orbit data.The algorithm uses third-degree spherical-radial cubature rule to approximate nonlinear equations’guess weighted integral which has higher estimation accuracy and numerical stability than Unscented Kalman Filter.J2 perturbation effects and atmosphere perturbation were considered to make LEO’s orbital dynamic model and the fourth order runge kutta method was used to discrete state equation in those ways the model precision has been improved.The orbit state equation and nonlinear velocity-data-measurement equation compose the system filtering equation.The simulation shows that the mean square error of the location is 7.059 m and the velocity estimation is 0.065 m/s when three stations are available.The results verifie that the method can satisfy the accuracy of LEO’s real-time orbit determination.