Adaptive Unscented Kalman Filter Based Estimation and Filtering for Dynamic Positioning with Model Uncertainties

A Framework of Cubature-Hᵢ/H∞-Fault Detection and Robust H-Infinity Kalman Filter of Ship SINS/GNSS Integrated System

“…In this section, the gain of the cubature-/ i HH  -FDF is adjusted in real time to enhance filter accuracy with k R -adaptation and k Q -adaptation, which can improve the final filter performance. Following the adaptive KF algorithm proposed in [44], the measurement noise covariance matrix k R -adaptation -FDF can be rewritten as follows: and an adaptive scaling factor can be introduced to adjust the process noise covariance k Q in real time. The algorithm is then presented with an adaptive scaling factor as follows:…”

Minimum error entropy based multiple model estimation for multisensor hybrid uncertain target tracking systems

“…Adaptive estimators: the adaptive estimation approach is mainly applied to the estimation of the unknown state and unknown noise parameter, which may considerably change over time in some cases. The adaptive estimators including the filtering tuning [11][12][13][14][15][16][17][18] and multiple model estimator (MM) [16,[19][20][21] are also the state estimation methods for the system uncertainties. There into, the uniqueness of the filtering tuning lies in adopting all kinds of adaptive tuning methods to suppress the system uncertainties, e.g.…”

“…There into, the uniqueness of the filtering tuning lies in adopting all kinds of adaptive tuning methods to suppress the system uncertainties, e.g. the noise tuning [11][12][13][14], the parameter tuning [15][16][17], and the switch-mode combination technique [18]. For example, an adaptive UKF (AUKF) was presented to simultaneously online adapt the process and measurement noise co-variances by adopting the main principle of the co-variance matching, which was applied to the ship dynamic positioning system with the model uncertainties of the time-varying noise statistics, the model mismatch, and the slow varying drift forces [11].…”

“…the noise tuning [11–14], the parameter tuning [15–17], and the switch‐mode combination technique [18]. For example, an adaptive UKF (AUKF) was presented to simultaneously online adapt the process and measurement noise co‐variances by adopting the main principle of the co‐variance matching, which was applied to the ship dynamic positioning system with the model uncertainties of the time‐varying noise statistics, the model mismatch, and the slow varying drift forces [11]. However, the AUKF can bring much cost burden due to the co‐variance matching, which does not benefit the real‐time of the system.…”

Variance-constrained $H_{\infty}$ filtering for multi-rate time-varying networked systems based on stochastic protocols

“…Filtering or state estimation problems have long been of far-reaching significance in signal processing, communication, control, and other fields and received many attentions in past few decades [1,2,8,18,22,31]. The celebrated Kalman filtering approach is one of the most popular ways to deal with the state estimation of a dynamic system [3,5,9]. It is noted that the design of the Kalman filter relies on the exact knowledge of the statistics of the external disturbance.…”