ISSN 2071-8594

Russian academy of sciences

Editor-in-Chief

Gennady Osipov

V. B. Melekhin, M. V. Khachumov Fuzzy Semantic Networks as an Adaptive Model of Knowledge Representation of Autonomous Intelligent Systems

Abstract.

The main features of planning the purposeful behavior of autonomous intelligent systems in conditions of a problem environment varying in degree of a priori uncertainty are considered. A model is developed for representing declarative knowledge of autonomous intelligent systems regardless of a specific subject area based on active and passive fuzzy semantic networks. The application of this model allows autonomous intelligent systems, on the one hand, to plan targeted behavior in a priori underdetermined conditions of a problem environment, and on the other hand, to organize the process of self-learning in a priori undescribed functioning conditions. The operations of comparing fuzzy semantic networks with each other are considered, which allow organizing effective decision-making in the process of planning targeted behavior in the face of uncertainty. The operations of decomposition, composition, and generalization of fuzzy semantic networks have been developed, which serve to organize the planning of the behavior of autonomous intelligent systems in the process of solving complex problems, accompanied by a formal description of current situations of a problem environment with a large dimension.

Keywords:

autonomous intelligent system, uncertainty conditions, problematic environment, knowledge representation, fuzzy semantic network, behavior planning.

DOI 10.14357/20718594200306

PP. 61-72.

References

1. Rassel, S., and P. Norvig. 2016. Iskusstvennyj intellekt: sovremennyj podhod [Artificial intelligence: a modern approach]. 2-e izd.: per.s angl. Moscow: Vil'yams. 1408 p.
2. Osipov, G.S. 2014. Lekcii po iskusstvennomu intellektu [Lectures on artificial intelligence]. Moscow: Liberkom. 272 p.
3. Lyugger, Dzh. F. 2005. Iskusstvennyj intellekt: strategii i metody resheniya slozhnyh problem [Artificial Intelligence: Strategies and Methods for Solving Complex Problems]: per. s angl. 4 – e izd. Moscjw: Vil'yams. 863 p.
4. Melekhib, V.B. 2019. Model of Representation and Acquisition of New Knowledge by an Autonomous Intelligent Robot Based on the Logic of Conditionally Dependent Predicates. Journal of Computer and Systems Sciences International. 58(5): 747 – 765.
5. Zade, L. 1976. Ponyatie lingvisticheskoj peremennoj i ego primenenie dlya prinyatiya: priblizhennyh reshenij [The concept of a linguistic variable and its application for adoption: approximate solutions]: per. s angl.. Moscow: Mir. 168 p.
6. Borisov, A.N. 1989. Obrabotka nechetkoj informacii v sistemah prinyatiya reshenij [Fuzzy information processing in decision making systems]. Moscow: Radio i svyaz'. 304 p.
7. Melekhin, V.B., and V.M. Hachumov. 2018. Upravlenie tekhnologicheskim processom s nechetkoj logikoj obrabotki znanij [Process control with fuzzy knowledge processing logic]. Pribory i sistemy. Upravlenie, kontrol', diagnostika [Devices and Systems. Management, control, diagnostics]. 2018. 5: 1 – 7.
8. Pegat A. 2013. Nechetkoe modelirovanie i upravlenie [Fuzzy modeling and control]: per. s angl. 2 – e izd. M.: Binom. 798 p.
9. Melekhin, V.B., and V.M. Hachumov. 2020. Upravlenie effektivnoj realizaciej tekhnologicheskih processov mekhanicheskoj obrabotki detalej v mashinostroenii [Management of the effective implementation of technological processes of machining parts in mechanical engineering]. Problemy upravleniya [Management Problems]. 1: 71 – 82.
10. Bershtejn, L.S., and V.B. Melekhin. 1988. Struktura procedurnogo predstavleniya znanij integral'nogo robota. 1. Rasplyvchatye semanticheskie seti [The structure of the procedural representation of knowledge of an integrated robot. 1. Vague semantic networks]. Izv. AN SSSR. Tekhn. Kibernetika [Izv. USSR Academy of Sciences. Tech. cybernetics]. 6: 119 – 124.
11. Bershtejn, L.S., and V.B. Melekhin. 1991. Dekompoziciya nechetkih semanticheskih setej dlya planirovaniya operacij integral'nogo robota [Decomposition of fuzzy semantic networks for planning operations of an integral robot]. Izv. RAN SSSR. Tekhn. Kibernetika [Izv. RAS of the USSR. Tech. cybernetics]. 1991. 5: 115 – 123.
12. Melekhin. V.B., and M.V. Hachumov. 2018. Planirovanie povedeniya intellektual'nogo bespilotnogo letatel'nogo apparata v nedoopredelennoj problemnoj srede. 1. Struktura i primenenie frejm-mikroprogramm povedeniya [Planning the behavior of an intelligent unmanned aerial vehicle in an undetermined problem environment. Part 1. Structure and application of frame-behavior microprograms]. Iskusstvennyj intellekt i prinyatie resheniya [Artificial Intelligence and Decision Making]. 2: 73 – 83.
13. Melekhin, V.B., and M,V. Hachumov. 2018. Planirovanie povedeniya intellektual'nogo bespilotnogo letatel'nogo apparata v nedoopredelennoj problemnoj srede. 2. Struktura i primenenie frejmov dejstvij [Planning the behavior of an intelligent unmanned aerial vehicle in an undetermined problem environment. Part 2. Structure and application of action frames]. Iskusstvennyj intellekt i prinyatie resheniya [Artificial Intelligence and Decision Making]. 2: 46 – 56.
14. Bershtein, L.S., and V.B. Melekhin. 1994. Planirovanie povedeniya intellektual'nogo robota [Intelligent robot behavior planning]. Moscow: Energoatomizdat. 240 p.
15. Bershtejn, L.S., and V.B. Melekhin. 1988. Struktura procedurnogo predstavleniya znanij integral'nogo robota. 2. Frejm mikroprogrammy povedeniya [The structure of the procedural representation of knowledge of an integrated robot. 2. The behavior of the firmware microprogram]. Izv. AN SSSR. Tekhn. Kibernetika [ Izv. USSR Academy of Sciences. Tech. cybernetics]. 5: 190 – 194.
16. Melikhov, A.N., L.S. Bershtejn, and S.Ya. Korovin. 1990. Situacionnye sovetuyushchie sistemy s nechetkoj logikoj [Fuzzy Situational Advisory Systems]. Moscow: Nauka. 272 p.