As a classical nonlinear filtering algorithm,converted measurement Kalman filter(CMKF)is com-monly employed to address the problem of target tracking when the measurements are in polar or spherical coor-dinates.Under the circumstance of low SNR,the mismatch between converted measurement distribution and co-variance matrix estimate will degrade radial estimation because of azimuth cosine nonlinearity.To solve nonlinear filtering problem with low SNR measurements,a mismatch factor was proposed to evaluate nonlinear influence,a decorrelated unbiased converted measurement(DUCM)model based on line-of-sight coordinate was developed,the weighted azimuth was introduced along visual axis to improve mismatch and down-range accuracy.Theoretic analysis and simulation verified that this decouple operation could improve radial estimation significantly,which could be applied to other converted measurement approaches.
关键词
视线坐标系/非线性滤波/去相关无偏转换量测/解耦算法
Key words
line of sight coordinate/nonlinear filtering/decorrelated unbiased conversion measurement/decou-pled algorithm
维普
万方数据