卡尔曼滤波的精度评定

Precision Evaluation of Kalman Filtering

卡尔曼滤波方法已广泛应用于动态测量数据的处理 ,其单位权方差通常按新息向量和观测值的残差向量进行计算 ,增加了观测值残差向量的计算 .本文根据新息向量和观测值残差向量与观测值的预测残差向量之间的关系 ,推导出用预测残差向量计算单位权的计算公式 。

Kalman filtering method is widely used in data adjustment of dynamic surveying system, whose unit variance is usually computed with innovation vector of parameters and residual vector of observations. According to the relationships of forecast residual vector between innovation vector of parameters and residual vector of observations, this paper deduces the formula of unit variance by forecast residual vector, which decreases calculation work.