The development of scientific and methodological provisions in the application of artificial neural networks

¹ Peter the Great St. Petersburg Polytechnic University, Saint-Petersburg, Russia
² Perm National Research Polytechnic University, Perm, Russia
³ Kazan National Research Technical University named after A. N. Tupolev-KAI, Kazan, Russia
⁴ Ferghana Polytechnic Institute, Ferghana, Uzbekistan

^* Corresponding author: pavel-mixailov1999@yandex.ru

Abstract

The relevance of the research topic in the field of artificial intelligence is associated with the growing interest in introducing independent intelligence through self-learning artificial neural networks, the need to expand the field of space research, reducing cosmonaut training and space objects research, as well as significant funds allocated to the subject under study. On the basis of the scientific researches to clarify the applied mathematical tools in the field of artificial intelligence and neural networks application; develop an algorithm for learning neural networks and identify the prospects for their use for artificial intelligence in space activities. In scientific work the following methods of scientific knowledge were used: analysis, synthesis, verification, deduction, modeling, etc. The algorithm for learning neural networks and a scheme for the principles of the development of artificial intelligence were compiled on the basis of the neuron activation functions. The studies of the authors are associated with the development of theoretical and methodological apparatus for the further economic and scientific-technological development of the production and economic system of Russia in the context of the introduction of artificial intelligence in the space industry, including both industrial and engineering parts of work, and theoretical physics.