Kautz-based adaptive control

The main purpose of this paper has been to investigate algorithms for adaptive control base on unstructured identification. Control low is a hybrid of two algorithms. Model reference and transfer function shaping are mixed and finally a new approach with all advantage of these two algorithms is obtained. Fundamental problems associated in adaptive control are: 1. The requirement of an exact prior system information such as degree and delay. 2. Sensitivity to variations of the system time delay. For solving these problems unstructured identification is offered. One of the most famous of OFS model is Kautz network that used. Particular systems can be well approximated by using Kautz model with complex poles. This method covers a wide range of stable linear systems, namely minimum phase, non-minimum phase and time delay systems.