Overview

One of the oldest and most fundamental of all human scientific pursuits is developing mathematical models for physical systems based on imperfect observations or measurements. This activity is known as system identification. This book describes the theory and practice of system identification applied to aircraft.

A good practical definition of system identification is that of Zadeh1:

System identification is the determination, on the basis of observation of input and output, of a system within a specified class of systems to which the system under test is equivalent.

Implicit in the preceding definition is the practical fact that the mathematical model of a physical system is not unique. In general, the guiding principle for model selection is the parsimony principle, which states that of all models in a specified class that exhibit the desired characteristics, the simplest one should be preferred. There are both theoretical and practical reasons for the parsimony principle, and these will be discussed further throughout the book. The preceding definition also mentions that system identification is based on observations of input and output for the system under test. In practice, these observations are corrupted by measurement noise. This requires the introduction of statistical theory and methods, as will be described in detail. Finally, there must also be some definition of what is meant by the word equivalent in the preceding definition. There is more than one way in which a model can be considered equivalent to a system under test. The most common approaches will be...