On the design of the feedforward compensator in two-degree-of-freedom controllers

This paper presents an innovative methodology to design the feedforward path of two-degree-of-freedom (2-d.o.f.) regulators for ‘optimal’ set point tracking. The proposed methodology is totally independent of the structure of the feedback controller, of the way it is designed, and of the characteristics of the set point signal. By using a nonparametric model of the control loop, that can be easily identified on-line, various tracking problems can be cast into a unified, and easily interpreted optimisation framework. Simulated and physical examples demonstrate the simplicity and efficacy of the proposal.