Identification of aircraft longitudinal stability and control derivatives combining global search natural algorithms and a gradient based algorithm

Abstract

In this work, two optimization methods are investigated to accomplish the parameter identification of the longitudinal motion of a military aircraft within the framework of the Output Error methodology. One of these methods is based on natural algorithms, in particular, the genetic algorithms and particle swarm optimization, which combined constitute the so called Life Cycle Method. The other is the Levenberg-Marquardt optimization algorithm, which is a gradient based method. Since the methodologies differ in the way they perform the optimization, being the first based on search and so appropriate for global minima search and the second gradient based, thus, good for local minima, both are compared and used in such a way that they complement each other. © 2009 IFAC.