Symbolic and numerical software tools for LFT-based low order uncertainty modeling

Abstract

One of the main difficulties in applying modern control theories for designing robust controllers for linear uncertain plants is the lack of adequate models describing structured physical model uncertainties. We present a systematic approach for the generation of uncerteinty models described by linear fractional transformations (LFTs) and report on recently developed symbolic and numerical software to assist the generation of low order LFT-based uncertainty models. The kernel of the symbolic software is a Maple library for generation and manipulation of LFT models. Additional numerical tools for order reduction of LFT models are based on MATLAB and FORTRAN implenmentations of numerically reliable algorithms. Three examples of uncertainty modeling of aircraft dynamics illustrate the capabilities of the new software to solve high order uncertainty modeling problems.