Model Error Quantification for Robust Control based on Quasi-Bayesian Estimation in Closed Loop

This paper presents a procedure for quantifying the errors in the estimation of the parameters of systems described by ARMAX models when operating in closed loop. We include stochastic disturbances on the output and consider the case where the true open loop plant is not a member of the chosen set of identifier models. This latter problem is dealt with by considering the impulse response of the undermodelling to be a particular realisation of a random vector with known characteristics but unknown parameters. We show how the parameters which characterize the undermodelling may be estimated from the data using Maximum Likelihood. For the case of Gaussian probability density functions we show how this information may be used to obtain a quasi-Bayesian estimate of the conditional distribution of the true system model. This in turn allows confidence regions to be established which would be suitable for use in robust control system design.

[1]  G. Goodwin,et al.  Quantification of Uncertainty in Estimation using an Embedding Principle , 1989 .

[2]  L. Valavani,et al.  A Frequency-Domain Estimator for Use in Adaptive Control Systems , 1987 .

[3]  David Q. Mayne,et al.  Optimal Non-stationary Estimation of the Parameters of a Linear System with Gaussian Inputs † , 1963 .

[4]  G. Goodwin,et al.  A stochastic embedding approach for quantifying uncertainty in the estimation of restricted complexity models , 1989 .

[5]  Michael Athans,et al.  A Frequency-Domain Estimator for Use in Adaptive Control Systems , 1987, 1987 American Control Conference.

[6]  Roy S. Smith,et al.  Towards a Methodology for Robust Parameter Identification , 1990, 1990 American Control Conference.

[7]  Lennart Ljung,et al.  On Estimation of Transfer Function Error Bounds , 1991 .

[8]  G. Goodwin,et al.  Optimal model order selection and estimation of model uncertainty for identification with finite data , 1991, [1991] Proceedings of the 30th IEEE Conference on Decision and Control.

[9]  Lennart Ljung,et al.  Influence of Model Order on Change Detection in Noise-Free, Complex System , 1990, 1990 American Control Conference.

[10]  R. Kosut.  Adaptive Robust Control via Transfer Function Uncertainty Estimation , 1988, 1988 American Control Conference.

[11]  R. Kosut.  Adaptive control via parameter set estimation , 1988 .

[12]  R. Younce,et al.  Identification with non-parametric uncertainty , 1990, IEEE International Symposium on Circuits and Systems.

[13]  P. Caines,et al.  Stationary linear and non-linear system identification and predictor set completeness , 1977, 1977 IEEE Conference on Decision and Control including the 16th Symposium on Adaptive Processes and A Special Symposium on Fuzzy Set Theory and Applications.

[14]  Graham C. Goodwin,et al.  Quantification of Uncertainty in Estimation , 1990, 1990 American Control Conference.

[15]  P. Caines.  Stationary linear and non-linear system identification and predictor set completeness , 1977 .

[16]  John B. Moore,et al.  Model approximations via prediction error identification , 1978, Autom..