Robust Extended Kalman Filtering for Nonlinear Systems With Stochastic Uncertainties

A Kalman Filter with Noise Covariances Estimation and Its Stability Analysis

“…However, in many practical systems, the noise covariance matrices are both completely unknown. Under this situation, using incorrect noise covariance matrices in Kalman filter may degrade the performance of the filter or even make the filter not work at all [15]. Therefore, it is necessary for a Kalman filter to obtain accurate noise covariances.…”

“…The main contribution of some works is to reduce the impact of the uncertainty of the noise covariances [15,16]. There are few algorithms were presented to estimate the covariance matrices of the process and measurement noises simultaneously.…”

“…In [14], a iterative adaptive unscented Kalman filter was proposed for sensor fusion to provide robustness with respect to the uncertainties in the system noise covariances. A robust extended Kalman filter was proposed in [15] for discrete-time nonlinear systems with stochastic uncertainties. The…”

Robust Kalman filtering for nonlinear systems with parameter uncertainties

“…It's well known that the uncertainties of the system degrade the performance of the KF based algorithms, and previous research have been improved the robustness of the filters [3][4][5]. These robust filters are less sensitive to deviations, but they require specific structures for the model uncertainties, so they are not easy to implement in practice.…”

EGMM: An enhanced Gaussian mixture model for detecting moving objects with intermittent stops

“…Then, we present our solution. In order to solve this problem, we not only combine the IGBM with GMM, but also incorporate an extended Kalman filter [6] based tracker into our scheme when the gray of this still object is similar to the gray of the background. In this way, we can retain the still object from moving state to still state alltime.…”