An evaluation of aircraft handling qualities by means of neural network based prediction model for characteristics of pilot control actions

Computer Science and Control


Аuthors

Tan W. , Efremov A. V.*, Tiumentsev Y. V.**

Moscow Aviation Institute (National Research University), 4, Volokolamskoe shosse, Moscow, А-80, GSP-3, 125993, Russia

*e-mail: pvl@mai.ru
**e-mail: yutium@gmail.com

Abstract

An approach is suggested to build a neural network based model, which allows to predict characteristics of pilot control actions. The model is developed using some data base containing experimental results. These results are obtained by means of test bench experiments involving human operator (pilot) for a set of different aircraft dynamical configurations. Frequency response of the neural network based model is compared with experimental results which was not utilized for a training of the network. The comparison demonstrates close fit for the theoretical and experimental data. The obtained model is used to derive selection criteria and handling qualities evaluation for manned aircraft.

mai.ru — informational site of MAI

Copyright © 1994-2024 by MAI