A sequential tracking filter without requirement of measurement decorrelation

A sequential measurement processing method in nonlinear state estimation is of benefit to both estimation accuracy and computational efficiency. When the measurement errors are correlated, the decorrelation is required to operate based on the covariance of the measurements in previous methods in order for correct sequential processing. In this paper, a new sequential processing method without decorrelation before filtering is presented. The measurements are processed directly by corresponding approaches, while the correlation is handled in the measurement update under the minimum mean squared error (MMSE) estimation framework. Simulations demonstrate the effectiveness of the new method. © 2012 ISIF (Intl Society of Information Fusi).