Overview

System identification, as it is termed today, is a scientific discipline that provides answers to the age-old inverse problem of obtaining a description in some suitable form for a system, given its behavior as a set of observations. The inverse problem and, hence, system identification has been a fundamental part of obtaining knowledge of any physical system that is observed, and as such has far ranging areas of applications. It enables us not only to know more about the principle underlying (i.e., model formulation of) the process being observed, but also to determine it in adequate detail (i.e., analysis and parameter estimation). System identification is implicitly concerned with modeling from experimental data, and covers applications in all possible areas: biology, medicine, chemical processes, economics, geology, materials, civil and mechanical engineering, automobiles, and, of course, flight vehicles, to name just a few. In this book we are concerned with the last-mentioned subject, flight vehicle system identification. As a rule, a flight vehicle is a dynamic system, and since a state space representation in the time-domain is closer to physical reality than any of the frequency-domain transform techniques, we deal here mainly with the time domain methodology.

The real processes are, in general, too complex and the exact internal behavior is unknown; the mathematical model, no matter how complex, at best represents a replica of what we purport to underlie the system behavior. It may be possible to postulate a very comprehensive model encompassing all conceivable influences or based on theoretical...