Keywords: robotics, emergency rescue operations, optimal control problem, maximum principle, artificial neural network training, safety
Neural network model of robotic complex control during emergency rescue operations under the conditions of the Far North
UDC 517.977.58
DOI: 10.26102/2310-6018/2023.40.1.010
The geopolitical situation and the increase in criminal and terrorist threats dictate the need to develop new territories, including the North of Russia. The article presents the results of a study of an artificial neural network mathematical model that takes into account delays designed to control an autopiloted ground-based robotic complex employed for transportation during emergency rescue operations under difficult natural and climatic conditions of the Arctic region. The universal nature of the proposed method is shown. An approach to finding a solution to the optimization problem using the necessary optimality conditions in the form of the Pontryagin maximum principle and the method of rapid automatic differentiation is described. A problem-oriented software product has been created, which is based on the developed computational algorithm for constructing approximate optimal control. The results obtained from numerical experiments confirm the effectiveness of the developed algorithm in finding an approximate optimal solution to the problem under consideration. The created software tool can be used to train an ANN with dynamics described by a system of differential equations with consideration to delays. The proposed mathematical model of ANN is suitable to solving a wide range of applied robotics tasks, including those aimed at developing technical means for emergency rescue operations under difficult natural and climatic conditions of the Arctic region. The flexibility, stability and adaptability of the selected model to changes in input parameters determine the prospects of using the developed computational algorithm to solve control problems in complex technical systems.
1. Bahtinova Ch.O., Chunaeva M.E. Avtomatizatsiya sistemy kontrolya kachestva pri organizatsii stroitel'stva osobo opasnykh i tekhnicheski slozhnykh ob"yektov v Rossii. Inzhenernyj vestnik Dona = Engineering journal of Don. 2022;3. (In Russ.). Available by: https://ivdon.ru/ru/magazine/archive/n3y2022/7511 (accessed on 26.11.2022).
2. Dushkin A.V., Cvetkov V.V. Voprosy modelirovaniya sostoyaniy inzhenerno-tekhnicheskikh sredstv okhrany i nadzora. Vestnik Voronezhskogo instituta FSIN Rossii = Vestnik of Voronezh Institute of the Russian Federal Penitentiary Service. 2014;3:28–31. (In Russ.).
3. Sumin V.I., Churakov D.yu., Car'kova E.G. Razrabotka modeley i algoritmov informatsionnykh struktur i protsessov ob"yektov osoboy vazhnosti. Promyshlennye ASU i kontrollery Industrial Automatic Control Systems and Controllers. 2019;4:30–39. (In Russ.).
4. Kayashev A.I., Rahman P.A., Sharipov M.I. Analiz pokazateley nadezhnosti lokal'nykh komp'yuternykh setey. Vestnik Ufimskogo gosudarstvennogo aviatsionnogo tekhnicheskogo universiteta. 2013;5:140–149. (In Russ.).
5. Bolodurina I.P., Ogurcova T.A., Arapova O.S., Ivanova Yu.P. Teoriya optimal'nogo upravleniya. Orenburg: Orenburgskij gosudarstvennyj universitet; 2016. 147 p. (In Russ.).
6. Tsarkova E., Belyaev A., Lagutin Y., Matveev Y., Andreeva E. Technical Diagnostics of Equipment Using Data Mining Technologies. Safety in Aviation and Space Technologies: Select Proceedings of the 9th World Congress «Aviation in the XXI Century». Cham: Springer. 2022:345–356. Available by: https://link.springer.com/chapter/10.1007/978-3-030-85057-9_30 (accessed on 26.11.2022).
7. Shanin D.A., Chikin V.V. Nejrosetevoj adaptivnyj kontroller dlya zadachi upravleniya ob’ektom s neizvestnoj strukturoj posredstvom global'noj obratnoj svyazi. Inzhenernyj vestnik Dona = Engineering journal of Don. 2008;2. (In Russ.). Available by: http://ivdon.ru/ru/magazine/archive/n2y2008/60 (accessed on 26.11.2022).
8. Evtushenko Yu.G. Metody resheniya ekstremal'nyh zadach i ih primenenie v sistemah optimizacii. Moskow; 1982. 432 p. (In Russ.).
9. Dushkin A.V., Kasatkina T.I., Novoseltsev V.I., Ivanov S.V. An improved method for predicting the evolution of the characteristic parameters of an information system. Journal of Physics: Conference Series. 2018;973(1):012031. Available by: https://iopscience.iop.org/article/10.1088/1742-6596/1003/1/012012/meta (accessed on 26.11.2022).
10. Dubrovin A.S., Ogorodnikova O.V., Tsarkova E.G., Andreeva E.A., Kulikova T.N. Analysis and visualization in graph database management systems. Journal of Physics: Conference Series: Current Problems. 2021;1902(1):012059. Available by: https://iopscience.iop.org/article/10.1088/1742-6596/1902/1/012059 (accessed on 26.11.2022).
Keywords: robotics, emergency rescue operations, optimal control problem, maximum principle, artificial neural network training, safety
For citation: Tsarkova E.G., Kalach A.V., Bobrov V.N. Neural network model of robotic complex control during emergency rescue operations under the conditions of the Far North. Modeling, Optimization and Information Technology. 2023;11(1). URL: https://moitvivt.ru/ru/journal/pdf?id=1291 DOI: 10.26102/2310-6018/2023.40.1.010 (In Russ).
Received 12.12.2022
Revised 27.01.2023
Accepted 10.02.2023
Published 31.03.2023