Nonlinear Kalman filtering using fuzzy local linear models

McGinnity, S.; Irwin, G.W.

doi:10.1109/acc.1997.612075

Cited by 3 publications

(2 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The fuzzy Kalman filter is designed based on the coverage of the state space by several local linear estimators according to the concept of fuzzy local linearization. It can be shown that the fuzzy Kalman filter can be written in the form of a Takagi-Sugeno-Kung (TSK) fuzzy model, which stands for local ARMAX models weighted by fuzzy membership functions [1][2][3][4]. The development of a method for validation of the fuzzy Kalman filter and for detecting incipient changes in its parameters, is important for a large number of applications in which such a type of distributed filtering is used (environmental surveillance, autonomous navigation systems, decentralized monitoring of industrial production etc.…”

Section: Introductionmentioning

confidence: 99%

Fuzzy Kalman Filter Validation Using the Local Statistical Approach

Rigatos

Siano

2015

Intell Ind Syst

View full text Add to dashboard Cite

In fuzzy Kalman filtering, each local Kalman filter makes use of its own model about the system's dynamics and the local state estimates are fused into one single state estimate through fuzzy weighting. These models can be inaccurate or the system's dynamics may change. The objective of the paper is to provide a method for finding out which one of the models used by the local Kalman filters contains parameters that deviate from the nominal values of the parameters of the real system. This is a problem of validation of the fuzzy Kalman filter, that is solved with the use of the local statistical approach to fault diagnosis. The development of a tool which can detect incipient changes in the parameters of the fuzzy Kalman filter is important, since such types of distributed Kalman filters are used in numerous applications that need precision and reliability in the provided state estimation (sensor networks, navigation systems, industrial production systems etc.).

show abstract

Section: Introductionmentioning

confidence: 99%

Fuzzy Kalman Filter Validation Using the Local Statistical Approach

Rigatos

Siano

2015

Intell Ind Syst

View full text Add to dashboard Cite

show abstract

“…The Kalman filter is also extensively used in combination with fuzzy systems. Since the Takagi-Sugeno fuzzy model (Wang et al, 2000) is a nonlinear combination of local linear models, Kalman filters have been used to develop state estimators for nonlinear systems which can be represented by models in this form (McGinnity and Irwin, 1997;Simon, 2003;Zhang and Wei, 2003). Fuzzy systems can also be used to tune parameters in a Kalman filter (Aja-Fernandez et al, 2003).…”

Section: Introductionmentioning

confidence: 99%

Distributed Kalman filtering for cascaded systems

Lendek

Babuška

Schutter

2008

Engineering Applications of Artificial Intelligence

View full text Add to dashboard Cite

The Kalman filter provides an efficient means to estimate the state of a linear process, so that it minimizes the mean of the squared estimation error. However, for naturally distributed applications, the construction and tuning of a centralized observer may present difficulties. Therefore, we propose the decomposition of a linear process model into a cascade of simpler subsystems and the use of a Kalman filter to individually estimate the states of these subsystems. Both a theoretical comparison and simulation examples are presented. The theoretical results show that the distributed observers, except for special cases, do not minimize the overall error covariance, and the distributed observer system is therefore suboptimal. However, in practice, the performance achieved by the cascaded observers is comparable and in certain cases even better than the performance of the centralized observer. A distributed observer system also leads to increased modularity, reduced complexity, and lower computational costs.

show abstract