针对卫星长期自主运行的发展要求,首次提出了基于滤波算法的卫星自主计算拟平均轨道根数的在轨实时方法。首先以受地球非球形摄动影响的低轨卫星为研究对象,推导了拟平均轨道根数的变化率并考虑了J2平方项补偿;以拟平均轨道根数为状态变量并以卫星瞬时速度和位置作为观测量建立了滤波方程,然后分别应用平方根无迹卡尔曼滤波(SR-UKF)和扩展卡尔曼滤波(EKF)对拟平均轨道根数进行估计。仿真结果表明,两种滤波方法均能有效地计算拟平均轨道根数,其中平方根UKF滤波具有更高的精度和稳定度,且都可以满足卫星在轨自主实时计算拟平均轨道根数的需求。
Aiming at real-time onboard estimation based on recursive fiher for autonomous satellite missions, a new method for extracting quasi-mean orbital elements from osculating orbital elements is proposed in this paper. Averaged variational equations of the geopotential of zonal harmonics and tesseral harmonies are derived to capture the evolution of the quasi-mean orbital elements. Considering the effects of J2 square term, the rates of change of quasi-mean orbital elements are formulated. Then state and measurement equations are formulated by treating the quasi-mean elements and osculating position and velocity as state variables and measurements respectively. The square root form of Unscented Kalman filter (SR-UKF) and extended Kalman filter (EKF) are utilized to estimate the quasi-mean elements. Finally, the simulation results prove that the proposed method is effective and has a high accuracy, which could satisfy the need of real-time onboard estimation for autonomous satellite.