A COMPARISON OF TWO ADAPTIVE ALGORITHMS FOR DISTURBANCE CANCELLATION

摘要

The objective of this paper is to compare two approaches to the adaptive rejection of disturbance inputs represented by a finite sum of sinusoids. The first approach is based on adapting the parameters of a finite impulse response (FIR) feedback controller using a least mean square (LMS) type algorithm to achieve asymptotic disturbance rejection. The second approach is based on using the recursive least squares (RLS) algorithm to search within the set of parametrized stabilizing controllers for a controller that leads to asymptotic disturbance rejection. The two approaches are compared in terms of their basic characteristics, as well as their performance in a simulation example.