基于平淡卡尔曼滤波器的微小卫星姿态确定算法

针对扩展卡尔曼滤波（EKF）在线性化过程中会引入误差的问题，采用平淡卡尔曼滤波器（UKF）进行了系统滤波器设计；提出一种构建虚拟观测量的方法，并分析了其噪声特性．虚拟观测量与高精度器件量测量搭配可实现对姿态的校正．以太阳敏感器、微电子机械系统（MEMS）陀螺、磁强计为姿态敏感器件，构建了定姿滤波器并用STK（Satellite Tool Kit）数据进行了仿真．结果表明，所提出方法能有效地提高定姿性能，采用UKF的系统定姿误差与EKF相当，但收敛时间、稳定性要优于EKF．

A Micro-Satellite Attitude Determination Algorithm Based on Unscented Kalman Filter

In view of the problem that the linearization process of extended Kalman filter(EKF) can introduce some errors into the system,the unscented Kalman filter(UKF) was utilized to design the system.A method of forming the virtual measurement was proposed,and its noise characters were analyzed.Combined with the high accuracy measurement,the virtual measurement can realize the measurement update.Taking sun sensor,magnetometer and MEMS gyro as attitude sensors,the attitude determination filter was constructed and simulation was proceeded using the STK(Satellite Tool Kit) data.The results show that the system performance is improved by the suggested method.The determination accuracy is comparable to EKF,while the converge speed and stability is higher.