Unscented Kalman Filter Using Augmented State in the Presence of Additive Noise
CASE '09 Proceedings of the 2009 IITA International Conference on Control, Automation and Systems Engineering (case 2009)
Extended Symmetric Sampling Strategy for Unscented Kalman Filter
CASE '09 Proceedings of the 2009 IITA International Conference on Control, Automation and Systems Engineering (case 2009)
Complexity analysis of the marginalized particle filter
IEEE Transactions on Signal Processing
Performance evaluation of UKF-based nonlinear filtering
Automatica (Journal of IFAC)
Hi-index | 0.01 |
Unscented Kalman filter (UKF) is a class of nonlinear filtering methods based on unscented transform within the Kalman filter framework. It is in light of the intuition that to approximate a probability distribution by a set of deterministic samples is easier than to approximate an arbitrary nonlinear transform. The key factors of UKF--the scalar, the state variable dimensions and the noises involved in nonlinear system, besides the probability distribution, should be synthetically analyzed. The mean square error is adopted to evaluate the effect of these factors on the performance of UKF. The simulation results show that the factors above mentioned more or less affect the performance of UKF, in which the state noise plays the most important role.