Optimum performance levels for minimax filters, predictors and smoothers
Systems & Control Letters
Kalman Filtering and Neural Networks
Kalman Filtering and Neural Networks
Estimation with Applications to Tracking and Navigation
Estimation with Applications to Tracking and Navigation
Optimal State Estimation: Kalman, H Infinity, and Nonlinear Approaches
Optimal State Estimation: Kalman, H Infinity, and Nonlinear Approaches
H∞ fixed-lag smoothing filter for scalarsystems
IEEE Transactions on Signal Processing
Least-squares estimation: from Gauss to Kalman
IEEE Spectrum
A view of three decades of linear filtering theory
IEEE Transactions on Information Theory
Hi-index | 0.08 |
The Kalman filter (KF) remains the most popular method for linear state and parameter estimation. Various forms of the KF have been created to handle nonlinear estimation problems, including the extended Kalman filter (EKF) and the unscented Kalman filter (UKF). The robustness and stability of the EKF and UKF can be improved by combining it with the recently proposed smooth variable structure filter (SVSF) concept. The SVSF is a predictor-corrector method based on sliding mode concepts, where the gain is calculated based on a switching surface. A phenomenon known as chattering is present in the SVSF, which may be used to determine changes in the system. In this paper, the concept of SVSF chattering is introduced and explained, and is used to determine the presence of modeling uncertainties. This knowledge is used to create combined filtering strategies in an effort to improve the overall accuracy and stability of the estimates. Simulations are performed to compare and demonstrate the accuracy, robustness, and stability of the Kalman-based filters and their combinations with the SVSF.