On-line Model-Free Methods

Abstract

Learning Objectives 3.1 Introduction 3.2 Iterative Feedback Tuning 3.3 The Controller Parameter Cycling Tuning Method 3.4 Summary and Future Directions = ∂ 2 ∂r ∂rAlgorithm 3.1 serves as a fundamental framework for this chapter, since in the sequel two methods are described for obtaining the gradient. The first is that of the Iterative Feedback Tuning technique and the second is the method of controller parameter cycling. Introduction 113

An Iterative Approach for Noncausal Feedforward Tuning

“…Remark 2. With respect to the classical deterministic version of the IFT method [20], here the considered cost function does not include the term related to the control effort (as in [19]). This is due to the fact that the trade-off between aggressiveness of the control system and its control effort can be easily handled by the choice of the transition time τ (see (3)) [13], [10].…”