Adaptive Inverse-lattice Learning Control

In this article the design framework for an Adaptive Inverse Lattice Controller(AILC) with learning attributes, applicable to linear Auto Regressive (AR) systems, is presented. The utilized controller structure relies on the principle of Inverse Model Control (IMC) and its topology resembles that of a lattice lter. The adaptation rules depend on the iden ti edsystem dynamics through an adaptive lattice lter. The identi cation scheme is extended with a proposed algorithm for the model order selection. Within the employed IMC{structure, an inverse lattice controller is utilized in the forward path in cascade with a lowpass detuning lter. As time progresses, the lattice lter estimates more accurately the system dynamics, and the learning scheme adjusts accordingly the attributes of the detuning lter. Simulation studies are used to in vestigate the eÆcacy of the suggested scheme. Copyright c 2002 IFA C