An adaptive-covariance-rank algorithm for the unscented Kalman filter

Lauren E. Padilla, Clarence W. Rowley
2010 49th IEEE Conference on Decision and Control (CDC)  
The Unscented Kalman Filter (UKF) is a nonlinear estimator that is particularly well suited for complex nonlinear systems. In the UKF, the error covariance is estimated by propagating forward a set of "sigma points," which sample the state space at intelligently chosen locations. However, the number of sigma points required scales linearly with the dimension of the system, so for large-dimensional systems such as weather models, the approach becomes intractable. This paper presents an
more » ... e version of the UKF, in which the error covariance is represented by a reduced-rank approximation, thereby substantially reducing the number of sigma points required. The method is demonstrated on a onedimensional atmospheric model known as the Lorenz 96 model, and the performance is shown to be close to that of a full-order UKF.
doi:10.1109/cdc.2010.5717549 dblp:conf/cdc/PadillaR10 fatcat:w5pcsvhb25ccboptvqkgxrwzim