Technical and biological parts of ergatic system "pilot-aircraft" accommodation using artificial neural network approach

Dynamics, ballistics, movement control of flying vehicles


Аuthors

Evdokimenkov V. N.*, Kim R. V.**, Yakimenko V. A.***

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

*e-mail: evdokimenkovvn@mai.ru
**e-mail: romanvkim@yandex.ru
***e-mail: whyacehka@gmail.com

Abstract

This paper presents an approach for operational control and post-flight analysis of pilot’s activity features, which would increase the level of safety and efficiency level of modern aircraft. The presented approach is based on usage of individual-adapted neural network models, which characterizes individual piloting style. The designed models might be used in so-called pilot’s support systems which main function is generating recommendations during the flight. These recommendations may increase the pilot’s control actions efficiency.

The considered models provide an opportunity for determining the prediction of accuracy of bringing the aircraft to the terminal point of the trajectory on the basis of the current ergatic system “pilot-aircraft” state, which includes components that describe technical state and pilot’s activity features. The structure, parameters and set of neural network models inputs are determined based on data recorded by flight data recorders in previous flights and continuously refined as new data become available. The multilayer perceptron (MLP) might be used as an example of considered artificial neural network model. The main property of MLP is an ability to approximate any functional relationship between current state of ergatic system “pilot-aircraft” and the prediction of accuracy.

We calculate the information signal value, which is displayed, on the basis of the difference between the predicted and required state in a terminal point. The given results confirm the efficiency of presented approach.

Keywords:

Artificial neural network, pilot’s activity model, pilot’s support systems

References

  1. Serebryakov G.G., Zheltov S.Y., Oginski A.A. Mashinostroenie. Enciklopediya. T.1-4. Avtomaticheskoe upravlenie. Teoriya. (Engineering. Encyclopedia. Vol. 1-4. Automatic control. Theory.), Mashinostroenie, 2000, pp. 604-641.

  2. Serebryakov G.G. Mechatronics, automation, control, 2010, no. 4, pp. 17-29.

  3. Serebryakov G.G. Vestnik komp’yuternykh i informatsionnykh tekhnologii, 2007, no. 10, pp. 2-8.

  4. Bazlev D.A., Evdokimenkov V.N., Kim N.V., Krasil’shchikov M.N. Vestnik komp’yuternykh i informatsionnykh tekhnologii, 2007, no.1, pp. 10-25.

  5. Bazlev D.A., Evdokimenkov V.N., Kim N.V., Krasil’shchikov M.N. Izvestiya RAN. Teoriya i sistemy upravleniya, 2008, no.4, pp. 97-108.

  6. Evdokimenkov V.N., Kim R.V., Krasil’shchikov M.N., Serebryakov G.G. Izvestiya RAN. Teoriya i sistemy upravleniya, 2015, no. 4, pp. 111-123.

  7. Evdokimenkov V.N., Kim R.V., Vekshina A.B., Yakimenko V.A. Vestnik Moskovskogo aviatsionnogo instituta, 2015, vol. 22, no. 3, pp. 17-29.

  8. Archipova O.V. Trudy MAI, 2012, no. 62: http://www.mai.ru/science/trudy/published.php?ID=35521


Download

mai.ru — informational site MAI

Copyright © 2000-2022 by MAI

Вход