O.N. Korsun – Dr. Sc. (Eng.) Professor, Head of Laboratory State Research Institute 

of Aviation Systems Scientific Center of Russian Federation (Moscow)

E-mail: marmotto@rambler.ru

P.I. Sotnikov – Post-graduate Student, 

Bauman Moscow State Technical University

E-mail: sotnikoffp@gmail.com

Assessment of the psychophysiological state of the human operator during the piloting is an important task in improving the airborne equipment of prospective aircrafts. One of the approaches to assessing the operator’s state, in particular, according to such symptoms as fatigue, falling asleep, etc., is the analysis of his electroencephalograms (EEG). In the future, monitoring of EEG during piloting will help to quickly detect and prevent undesired operator states.

The paper investigates the possibility of assessing the quality of the flight task performance based on the analysis of the operator's electroencephalogram (EEG). We conducted an experiment in which operator carried out multiple landing approaches at the flight simulator for an hour. Simultaneously we recorded EEG using a 14-channel wireless Emotiv Epoc encephalograph. After completing the experiment, we divided landing approaches into "good" and "bad" ones depending on the value of the standard deviation of the altitude error (vertical deviation from the glide path). Then we solved the binary classification problem of EEG records belonging to the "good" and "bad" landing approaches. At the first stage, for all EEG segments we formed vectors of characteristic features. At the second stage, we used the obtained feature vectors for training and testing classifiers of different types. It was found that the probability of correct recognition of the landing approach type significantly exceeds the probability of random classification (50%) for all EEG feature extraction methods and classification methods selected for comparison. The maximum classification accuracy (87%) was achieved by combining the feature extraction method based on optimal frequency ranges search and the ν-SVM classifier. In addition, we performed a shapelets search among all EEG signal spectra. The shapelets search showed that "bad" landing approaches (with a high vertical deviation from the glide path) are distinguished by increased theta rhythm power in frontal areas. The obtained results indicate that EEG can be used to control the quality of flight tasks performed by a human operator.

Korsun O.N., Sotnikov P.I. Investigation of possibilities for the monitoring flight tasks performance using the electroencephalogram of a human operator. Highly Available Systems. 2020. V. 16. № 2. P. 33–43. DOI: 10.18127/j20729472-202002-03 (In Russian).

1. Polikanova I.S., Leonov S.V. Psihofiziologicheskie i molekulyarno-geneticheskie korrelyaty utomleniya,. Elektronnyj zhurnal «Sovremennaya zarubezhnaya psihologiya». 2016. T. 5. № 4. S. 24–35 (In Russian).
2. Craig A., Tran Y., Wijesuriya N., Nguyen H. Regional brain wave activity changes associated with fatigue. Psychophysiology. April 2012. V. 49. № 4. P. 574–82.
3. Polikanova I.S. Sergeev A.V. Vliyanie dlitel'noj kognitivnoj nagruzki na parametry EEG. Nacional'nyj psihologicheskij zhurnal. 2014. № 1 (13). S. 84–92 (In Russian).
4. Trejo L.J., Kochavia R., Kubitz K., Montgomery L.D., Rosipal R., and Matthews B. EEG-based Estimation of Cognitive Fatigue. Proceedings of Symposium OR05 Defense and Security. 2005. V. 5797. P. 105–115.
5. Lal S., Bekiaris E. The Reliability of Sensing Fatigue from Neurophysiology. AusWireless 2006: International Conference on Wireless Broadband and Ultra Wideband Communications Proceedings. Sydney. 2007. P. 1–4.
6. Guseva N.L., Sofronov G.A., Suvorov N.B. Osobennosti dinamiki al'fa-ritma elektroencefalogrammy i kardioritmogrammy cheloveka pri snizhenii urovnya bodrstvovaniya. Vestnik rossijskoj voenno-medicinskoj akademii. 2007. № 3 (19). S. 24–31 (In Russian).
7. Korsun O.N., Nabatchikov A.M., Burlak E.A. Sinhronizaciya informacionnyh potokov pri polunaturnom modelirovanii dvizheniya letatel'nyh apparatov. Elektronnyj nauchno-tekhnicheskij zhurnal «Inzhenernyj vestnik». 2013. № 10. S. 1–16 (In Russian).
8. Levickaya O.S., Lebedev M.A. Interfejs mozg-komp'yuter: budushchee v nastoyashchem. Vestnik RGMU. 2016. № 2. S. 4–16 (In Russian).
9. Lotte F., Congedo M., Lécuyer A., Lamarche F. A review of classification algorithms for EEG-based brain–computer interfaces. Journal of Neural Engineering, 2007. V. 4. № 2. P. 24.
10. Tereshchenko E.P., Ponomarev V.A., Kropotov Yu.D., Myuller A. Sravnenie effektivnosti razlichnyh metodov udaleniya artefaktov morganij pri analize kolichestvennoj elektroencefalogrammy i vyzvannyh potencialov. Fiziologiya cheloveka. 2009. T. 35. № 2. S. 124–131 (In Russian)
11. Korsun O.N., Mihajlov E.I. Metody analiza elektroencefalogramm v celyah ocenki sostoyaniya cheloveka-operatora v processe pilotirovaniya. Cloud of Science. 2018. T. 5. № 4. S. 649 – 663 (In Russian).
12. Mikhaylov E.I., Korsun O.N. Algorithms of the operator's electroencephalogram analysis based on the principal component analysis. ICPE 2018 – Int. Conf. on Psychology and Education: The European Proceedings of Social & Behavioural Sciences EpSBS. 2018. P. 445 – 450.
13. Sotnikov P.I. Vydelenie harakternyh priznakov signala elektroencefalogrammy s pomoshch'yu analiza entropii. Nauka i Obrazovanie. MGTU im. N.E. Baumana. Elektronnyj zhurnal. Noyabr' 2014. № 11. S. 555–570 (In Russian).
14. Blankertz B., Tomioka R., Lemm S., Optimizing Spatial Filters for Robust EEG Single-Trial Analysis. IEEE Signal Processing Magazine. 2008. V. XX. P. 581–607.
15. Rejer I. Genetic Algorithms in EEG Feature Selection for the Classification of Movements of the Left and Right Hand. Proceedings of the 8th International Conference on Computer Recognition Systems CORES 2013. Advances in Intelligent Systems and Computing. 2013. V. 226. P. 579–589.
16. Sotnikov P.I. Metody postroeniya prostranstva priznakov signala EEG v gibridnom interfejse «glaz-mozg-komp'yuter». Matematika i matematicheskoe modelirovanie. 2018. № 2. S. 33–52 (In Russian).
17. Oana D.E., Anca M.L. Comparison of Classifiers and Statistical Analysis for EEG Signals Used in Brain Computer Interface Motor Task Paradigm. International Journal of Advanced Research in Artificial Intelligence. 2015. V. 4. № 1. P. 8–12.
18. Muller K.R., Anderson C.W., Birch G.E. Linear and nonlinear methods for brain-computer interfaces. IEEE Transactions on Neural Systems and Rehabilitation Engineering. 2003. V. 11. № 2. P. 165–169.
19. Karpenko A.P., Kostrubin M.S., Chernyshev A.S. Effektivnost' klassifikacii mnogomernyh vremennyh ryadov s pomoshch'yu shejpletov. Nauka i Obrazovanie. MGTU im. N.E. Baumana. Elektronnyj zhurnal. 2015. № 11. S. 382–405 (In Russian).