Particle filtering‐based recursive identification for controlled auto‐regressive systems with quantised output

Intl J Robust & Nonlinear

2020

122

Summary This paper is concerned with the design of a state filter for a time‐delay state‐space system with unknown parameters from noisy observation information. The key is to investigate new identification algorithms for interactive state and parameter estimation of the considered system. Firstly, an observability canonical state‐space model is derived from the original model by linear transformation for the purpose of simplifying the model structure. Secondly, a direct state filter is formulated by minimizing the state estimation error covariance matrix on the basis of the Kalman filtering principle. Thirdly, once the unknown states are estimated, a state filter–based recursive least squares algorithm is proposed for parameter estimation using the least squares principle. Then, a state filter–based hierarchical least squares algorithm is derived by decomposing the original system into several subsystems for improving the computational efficiency. Finally, the numerical examples illustrate the effectiveness and robustness of the proposed algorithms.

Section: The Problem Formulationmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

unclassified

See 1 more Smart Citation

Adaptive parameter estimation for a general dynamical system with unknown states

Xiao

Intl J Robust & Nonlinear

2020

122

“…Recently, a switched gradient descent algorithm was proposed to solve the problem in estimating the secure state of the cyber‐physical systems . Furthermore, the parameter identification issues of controlled auto regressive systems with quantized output were investigated by integrating the auxiliary model with the gradient search . Some parameter estimation methods can be used for establishing the models of different plants and engineering systems .…”

Section: Introductionmentioning

confidence: 99%

“…23 Furthermore, the parameter identification issues of controlled auto regressive systems with quantized output were investigated by integrating the auxiliary model with the gradient search. 24 Some parameter estimation methods can be used for establishing the models of different plants and engineering systems. [25][26][27][28] The innovation is the useful information which can be utilized for improving the parameter estimation accuracy.…”

mentioning

confidence: 99%

Recursive methods for estimating the radial basis function‐based state‐dependent autoregressive model

Zhou

Intl J Robust & Nonlinear

et al. 2020

Summary Identifying a nonlinear radial basis function‐based state‐dependent autoregressive (RBF‐AR) time series model is the basis for solving the corresponding prediction and control problems. This paper studies some recursive parameter estimation algorithms for the RBF‐AR model. Considering the difficulty of the nonlinear optimal problem arising in estimating the RBF‐AR model, an overall forgetting gradient algorithm is deduced based on the negative gradient search. A numerical method with a forgetting factor is provided to solve the problem of determining the optimal convergence factor. In order to improve the parameter estimation accuracy, the multi‐innovation identification theory is applied to develop an overall multi‐innovation forgetting gradient (O‐MIFG) algorithm. The simulation results indicate that the estimation model based on the O‐MIFG algorithm can capture the dynamics of the RBF‐AR model very well.

The innovation algorithms for multivariable state‐space models

Adaptive Control & Signal

Xiao

2019

115

This paper derives the input-output representation of the dynamical system described by a linear multivariable state-space model and the corresponding multivariate linear regressive model (ie, multivariate equation-error model). A projection identification algorithm, a multivariate stochastic gradient identification algorithm, and a multi-innovation stochastic gradient (MISG) identification algorithm are proposed for multivariate equation-error systems by using the negative gradient search and the multi-innovation identification theory. The convergence analysis of the MISG algorithm indicates that the parameter estimation errors converge to zero under the persistent excitation condition. Finally, a numerical example illustrates the effectiveness of the proposed algorithms.