Robust multivariable implicit self-tuning control

Qufei Song, M.R. Katebi, M.J. Grimble

IMA Journal of Mathematical Control and Information · 1993 · 19 citations · 0 references

Concepts

Abstract

A robust multivariable linear quadratic implicit self-tuning algorithm is presented. It is shown that the closed-loop output error is bounded for multiplicative modelling uncertainty, provided that all the unstable zeros of the plant are captured in the nominal model. The algorithm can be applied to nonminimum-phase systems, and does not require the explicit solution of the algebraic matrix polynomial equation. The controller parameters are directly estimated from two implicit prediction models which contain the plant and controller parameters in bilinear form. The stability analysis is performed by applying conic-sector theory to a new error equation which is decomposed into two components embodying a linear and non-linear operator, namely the modelling uncertainty and the parameter estimator respectively. Simulation results are presented to demonstrate the performance of the proposed adaptive controller.