自适应扩展卡尔曼滤波在卫星姿态确定系统中的应用

Application of adaptive extended Kalman filter in spacecraft attitude determination system

扩展卡尔曼滤波(以下简称EKF)算法应用于卫星姿态确定系统时需要已知精确的系统模型及过程噪声和观测噪声统计特性,并有计算量过大的问题。本文在EKF算法中加入噪声观测器,构成自适应扩展卡尔曼滤波算法(Adaptive Extended Kalman Filter,以下简称AEKF),使系统能够在传感器噪声统计特性未知的情况下,依然获得较高的系统状态估计精度,增强了系统的鲁棒性。并且AEKF算法简化了系统状态方程,相对于EKF算法减少计算量。经数学仿真验证,AEKF算法能较好地对传感器噪声的统计特性进行在线估计,使姿态确定系统正常工作,有较高的工程应用价值。

It is well known that the main drawbacks of traditional extended Kalman filter （EKF） requires accurate system model, statistics properties of process noise and observation noise and great amount computation. To deal with the problems,an adaptive extended Kalman filter （AEKF） with noise observer was introduced in the spacecraft attitude determination system. This AEKF can obtain high estimation accuracy of system state without sensor noise statistics and improve system robustness. It can also simplify system state model from 6-D to 3-D so that the computation burden was largely reduced. Simulation results indicate that AEKF algorithm can estimate sensor noise statistics on line and make attitude determination system work well.