CLUSTER ANALYSIS OF A COLLECTIVE OF ALGORITHMS FOR MULTICORE NEURAL NETWORK AUTOMATES AND ROBOTS ON CHIP
V. N. Ruchkin, V. A. Fulin, E. V. Ruchkina, D. V. Grigorenko Ryazan State University named after S.A. Yesenin, Ryazan State Agrotechnological University named after P. A. Kostychev, CJSC "Ryazanpribor"
Annotation: Within the framework of increasing the efficiency of new spheres and directions of development of society, the state pays attention to robotization on a modern domestic basis in order to implement import substitution. One of the urgent problems is the combination of the concepts of a collective of algorithms, a collective of automata, a collective of robots and artificial intelligence. A special role is played by the possibilities of cybernetic research of multicore neural network automata in order to build more complex automata, robots and the behavior of a team of robots based on them. The purpose of this article is to demonstrate the possibilities of a set–theoretic approach of a cybernetic approach to artificial, complex natural objects and systems on this basis and to create a conceptual model for the selection and joint simultaneous design of hardware and software of neural network automata based on a unified study of the processes of parallelization of a collective of algorithms in the form of explicit and implicit clustering. As a result, the authors analyze, show and propose variants of the collective structures of algorithms for ensuring cybersecurity and protection against threats in the form of a hierarchy of security practices. The method of analysis and selection of the best architecture of a multicore neural network collective of an automaton and a robot collective based on automata implemented on a chip is proposed. An expert system based on VLSI 1879VM8YA (NM6408) with a developed user interface is being implemented.
Keywords: multiple-theoretic approach, team of algorithms, cybernetic research, team of robots, conceptual model, hardware and software design, explicit and implicit clustering, cybersecurity, threat protection, expert system, NM Card tool module, user interface.