Recursive Bayesian estimation using piece-wise constant approximations
Automatica (Journal of IFAC)
Posterior Cramer-Rao bounds for discrete-time nonlinear filtering
IEEE Transactions on Signal Processing
An extended Kalman filter frequency tracker for high-noiseenvironments
IEEE Transactions on Signal Processing
A tutorial on particle filters for online nonlinear/non-GaussianBayesian tracking
IEEE Transactions on Signal Processing
The extended Kalman filter as an exponential observer for nonlinearsystems
IEEE Transactions on Signal Processing
Correspondence: Comments on "Performance evaluation of UKF-based nonlinear filtering"
Automatica (Journal of IFAC)
Automatica (Journal of IFAC)
A new autonomous celestial navigation method for the lunar rover
Robotics and Autonomous Systems
Brief paper: Adaptive divided difference filtering for simultaneous state and parameter estimation
Automatica (Journal of IFAC)
Brief paper: Derivative-free estimation methods: New results and performance analysis
Automatica (Journal of IFAC)
Performance analysis of UKF for nonlinear problems
IITA'09 Proceedings of the 3rd international conference on Intelligent information technology application
SINS/CNS integrated navigation solution using adaptive unscented Kalman filtering
International Journal of Computer Applications in Technology
Automatica (Journal of IFAC)
A Gaussian approximation recursive filter for nonlinear systems with correlated noises
Automatica (Journal of IFAC)
TDOA-based adaptive sensing in multi-agent cooperative target tracking
Signal Processing
Hi-index | 22.16 |
The performance of the modified unscented Kalman filter (UKF) for nonlinear stochastic discrete-time system with linear measurement equation is investigated. It is proved that under certain conditions, the estimation error of the UKF remains bounded. Furthermore, it is shown that the design of noise covariance matrix plays an important role in improving the stability of the algorithm. Error behavior of the UKF is then derived in terms of mean square error (MSE), and the Cramer-Rao lower bound (CRLB) is introduced as a performance measure. The modified UKF is found to approach the CRLB if the difference between the real noise covariance matrix and the selected one is small enough. These results are verified by using Monte Carlo simulations on two example systems.