摘要
从卡尔曼滤波的新息出发,推导了卡尔曼滤波残差的新息表达式,并用其构造抗差因子.根据精密单点定位的内部可靠性和观测值的随机性,确定完全可探测的载波相位和伪距粗差,并分别加入相应观测值粗差.通过标准卡尔曼滤波、新息的抗差卡尔曼滤波和残差的抗差卡尔曼滤波3种方案对比,粗差使标准卡尔曼滤波的估计结果产生一定的偏差.当观测值精度不等时,新息的抗差卡尔曼滤波对较高精度观测值的粗差抗差效果较差.而残差的抗差卡尔曼滤波能够实现不等精度观测值粗差的同时抗差.
Residuals of Kalman filtering is derived and expressed by innovation and then used as robust factor for robust Kalman filtering. Then, according to the internal reliability of precise point positioning and random of measurements, the outliers absolutely detected are added to both carrier phase and pseudorange. Finally, three schemes including Kalman filtering, inno- vation-based robust Kalman filtering and residuals-based robust Kalman filtering are compared. The results show that there are biases in the estimated coordinates of standard Kalman filtering due to the effects of the added outliers. Innovation-based robust Kalman filtering is only able to reject the outliers of the observation with lower precision when the observations are with differ- ent precisioin. Residuals-based robust Kalman filtering has the ability of robust for the obser- vation with different precision.
出处
《中国矿业大学学报》
EI
CAS
CSCD
北大核心
2012年第5期857-862,共6页
Journal of China University of Mining & Technology
基金
国家自然科学青年基金项目(40904004)
国家自然科学基金项目(41074010)
关键词
抗差卡尔曼滤波
新息
残差
内部可靠性
粗差
robust Kalman filter
innovation
residuals
internal reliability
outliers
