D.V. Aladin1
1 Bauman Moscow State Technical University (Moscow, Russia)

1 aladin.dv@yandex.ru

In the course of research, a model has been created to solve management tasks in state space, which was used in systems that automatically generate Mivar knowledge bases. The research used the theoretical foundations of Mivar-based approach, including concepts such as Mivar space and principles for building Mivar networks. Special attention has been paid to methods of formalizing knowledge in the form of Mivar networks and dividing tasks into subtasks. A detailed structure of the model has been presented, which included elements for converting incoming data, preparing management decisions, and generating management signals. Mechanisms for building the Mivar knowledge base have been described, as were features of working in state space. Methods for automatically generating Mivar knowledge bases for robotic control systems and complexes have been also discussed.

The model has a number of significant advantages, including flexibility of application, versatility of structure and scalability of solution. The practical significance of this model is confirmed by its adaptability to different subject areas. An important advantage of this model lies in its ability to scale up and reuse formalized knowledge. Automatic generation of knowledge bases based on results from system analysis of subject areas significantly simplifies the process of solving typical tasks.

This developed model is a valuable tool for creating advanced Mivar expert systems with automatic generation, capable of handling complex management tasks across various subject areas. Future development of this model could include expansion into additional application areas and enhancement of automatic generation method.

Aladin D.V. Mivar knowledge base model for management task solving in state space. Neurocomputers. 2025. V. 27. № 4. P. 66–71. DOI: https://doi.org/10.18127/j19998554-202504-06 (in Russian)

1. Varlamov O.O. Wi!Mi expert system shell as the novel tool for building knowledge-based systems with linear computational complexity. International Review of Automatic Control. 2018. V. 11. № 6. P. 314–325. DOI: 10.15866/ireaco.v11i6.15855. (in Russian)
2. Varlamov O., Aladin D. A new generation of rules-based approach: Mivar-based intelligent planning of robot actions (MIPRA) and brains for autonomous robots. Machine Intelligence Research. 2024. V. 21. № 5. P. 919–940. DOI: 10.1007/s11633-023-1473-1.
3. Aladin D.V., Aladina E.V., Chuvikov D.A., Varlamov O.O., Adamova L.E. Creating a «Logical intelligent plant care system» in digital agriculture based on Mivar approach. IOP Conference Series: Earth and Environmental Science. 2022. P. 012004. DOI: 10.1088/1755-1315/954/1/012004.
4. Varlamov O.O. Osnovy sozdaniya mivarnykh ekspertnykh sistem. M.: INFRA-M. 2021. (in Russian)
5. Varlamov O.O. Mivarnye bazy dannykh i pravil. M.: INFRA-M. 2021. (in Russian)
6. Bolotova L.S. Sistemy iskusstvennogo intellekta: modeli i tekhnologii, osnovannye na znaniyakh. M.: Finansy i statistika. 2012. (in Russian)
7. Varlamov O.O. O trebovaniyakh k podgotovke iskhodnykh dannykh dlya mivarnykh sistem prinyatiya reshenij robotov. Sb. statej II Vseros. nauch. konf. «Iskusstvennyj intellekt v avtomatizirovannykh sistemakh upravleniya i obrabotki dannykh». Moskva, 27–28 aprelya 2023 g. M.: Izdatel'skij dom KDU, Dobrosvet. 2023. T. 3. S. 15–20. (in Russian)
8. Varlamov O.O. O metrike avtonomnosti i intellektual'nosti robototekhnicheskikh kompleksov i kiberfizicheskikh system. Radiopromyshlen­nost'. 2018. № 1. S. 74–86. (in Russian)
9. Popova I.A., Afanas'ev G.I., Timofeev V.B., Gapanyuk Yu.E. Issledovanie metodov mashinnogo obucheniya bez uchitelya dlya analiza i podgotovki dannykh i sovremennykh podkhodov k postroeniyu modelej mashinnogo obucheniya – sistem avtomatizirovannogo mashin­nogo obucheniya. Nejrokomp'yutery: razrabotka, primenenie. 2025. T. 27. № 1. S. 47–55. DOI: 10.18127/j19998554-202501-04.