System Identification: Theory for the User, 2nd Edition

11. Recursive Estimation Methods

11.1 Introduction

In many cases it is necessary, or useful, to have a model of the system available on-line while the system is in operation. The model should then be based on observations up to the current time. The need for such an on-line model construction typically arises since a model is required in order to take some decision about the system. This could be

• Which input should be applied at the next sampling instant?

• How should the parameters of a matched filter be tuned?

• What are the best predictions of the next few outputs?

• Has a failure occurred and, if so, of what type?

Methods coping with such problems using an on-line adjusted model of some sort are usually called adaptive (see Figure 11.1 ...