Computer Science and Control
Аuthors
, *, **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 |