Get the latest Science News and Discoveries
Accuracy evaluation of marginalized unscented Kalman filter - EurekAlert
State estimation has a variety of applications in engineering fields such as target tracking and spacecraft navigation. For linear models with Gaussian white noises, the Kalman filter (KF) exactly presents the optimal solution from the perspective of the minimum mean square error estimation. Plenty of effective suboptimal filtering algorithms have been developed for nonlinear cases, one of which is the unscented Kalman filter (UKF). The core of UKF is to calculate the values of state and measurement functions at a set of deterministically placed sampling points (i.e., sigma points) and to use weighted sums to compute the time and measurement updated means and covariances, guaranteeing the second-order accuracy in mean and covariance estimates. Compared with a conventional UKF, the recently proposed marginalized unscented Kalman filter (MUKF) uses a partially sampling strategy to achieve similar filter accuracy with fewer sigma points, demonstrating its powerful ability to deal with state estimation with mixed linearity and nonlinearity. However, the hypothesis that the accuracy of MUKF is equivalent to that of UKF is not true for all systems. In a research article recently published in Space: Science & Technology, scholars from Beijing Institute of Control Engineering and Beihang University evaluate the accuracy of MUKF, providing references for engineers to choose MUKF or UKF in different systems.
None
Or read this on Eureka Alert