Causal design methodology for optimal tracking in 2-d.o.f. industrial controllers

This paper presents and discusses a causal method to design the feedforward part of 2-d.o.f. industrial controllers for optimal tracking, independently of the structure of the feedback controller, of the way it was designed, and of the characteristics of the set point signal. The method is particularly well suited for embedded systems, or wherever the limits imposed by the hardware and/or software architecture play a significant role; it is based on a nonparametric model of the control loop, that can be easily identified on-line, and allows to cast various tracking problems into a unified optimization framework. An experimental application example demonstrates the method’s effectiveness, simplicity, and practical usefulness.