Date of Award

12-1994

Document Type

Thesis

Degree Name

Master of Science

Department

Department of Aeronautics and Astronautics

First Advisor

Stuart Kramer, PhD

Abstract

Gain scheduling. the traditional method of providing adaptive control to a nonlinear system, has long been an ad hoc design process. Until recently; little theoretical guidance directed this practitioners' art. For this reason a systematic study of this design process and its potential for optimization has never been accomplished. Additionally, the nonlinearities and the large search space involved in gain scheduling also precluded such an optimization study. Traditionally, the gain scheduling process has been some variation of a linear interpolation between discrete design points. By using powerful non-traditional optimization tools such as genetic algorithms there are ways of improving this design process. This thesis utilizes the power of genetic algorithms to optimally design a gain schedule. First, a design methodology is validated on a simple pole placement problem, then demonstrated for an F-18 Super-maneuverable Fighter. From this experience, a general gain scheduling design process is developed and presented.

AFIT Designator

AFIT-GAE-ENY-94D-3

DTIC Accession Number

ADA289306

Share

COinS