A. A. Zhdanov – Dr.Sc. (Phys.-Math.), Professor, Chief Research Scientist, Department of Scientific Preparation of Projects, Institute of Precision Mechanics and Computer Engineering n.a. S.A. Lebedev of the Russian Academy of Sciences; Professor, Basic Department of Computer Engineering, Faculty of Radio Engineering and Cybernetics, Moscow Institute of Physics and Technology

E-mail: a.zhdanov@mail.ru

V. M. Lazarev – Dr.Sc. (Eng.), Professor, Head of Management, JSC “Control Systems” (Moscow)

E-mail: lazarev@oaosu.ru

R. E. Peshenko – General Director, Research and Development Company “Intellect” JSC (Moscow)

E-mail: romanstep66@gmail.com

Effective management of modern and widely used in industrial practice of various industries groupings of robotic complexes presupposes a rational combination of the principles of centralization and decentralization together with the introduction of elements of artificial intelligence into the control loops. The purpose of the robotic systems control complex (RTC) can be associated with geological exploration of minerals, monitoring of oil and gas pipelines, flooding zones and fires, traffic flows (logistics issues), and many other applications. The key technology for the development of robotic systems should be considered the development of intelligent components of control systems (CS), ensuring the maximum possible autonomy of the RTC. At the same time, a breakthrough technology should be the creation of biologically-inspired control systems with adaptive (self-learning) properties that ensure the survival and self-development of all natural control systems.

Modern control systems for robots are built on basis of the deterministic methods of automatic control or using the methods of machine learning with a priori learning. But real conditions of the robots using demand real adaptive control where learning and control perform in one real-time process. The aim of the work is to analyze the mechanisms of centralized and decentralized control of robotic complexes group.

Approaches to the construction of control systems for robotic complexes, their advantages and disadvantages have been shown. The necessity of using adaptive control systems has been substantiated and their possibilities of overcoming the indicated disadvantages have been noted. Methods of adaptive control that can be applied in the described cases have been considered. The proposed methods of autonomous adaptive controls may be widely used in all levels of control of robots include groups of robots.

Zhdanov A.A., Lazarev V.M., Peshenko R.E. Mechanisms of centralized and decentralized control of robotic complexes group. Neurocomputers. 2020. Vol. 22. No. 4. P. 28–37. DOI: 10.18127/j19998554-202004-05. (in Russian)

1. Zhdanov A.A. Avtonomnyj iskusstvennyj intellekt. Izd. 2-e. M.: BINOM. Laboratoriya znanij. 2009. (in Russian)
2. Kalyaev I.A. Strategii gruppovogo upravleniya. Doklad v RAN. 2014. (in Russian)
3. Kalyaev I.A., Kapustyan S.G., Gajduk A.R. Samoorganizuyushchiesya raspredelennye sistemy upravleniya gruppami intellektual'nykh robotov, postroennykh na osnove setevoj modeli. Upravlenie bol'shimi sistemami. 2010. № 30-1. S. 605–639. (in Russian)
4. Dubin A.E., Lazarev V.M., Sviridov V.V. Avtonomnye transportnye sistemy i mobil'nye robotizirovannye platformy. Serpukhov (FVA RVSN). 2019. (in Russian)
5. Pshikhopov V.Kh., Medvedev M.Yu. Otsenivanie i upravlenie v slozhnykh dinamicheskikh sistemakh. M.: Fizmatlit. 2009. (in Russian)
6. Sheremet I.B., Rudianov N.A., Ryabov A.V., Khrushchev B.C., Komchenkov V.I. Obosnovanie semejstva boevykh i obespechivayushchikh robotov dlya boya v gorode. Izvestiya YuFU. Tekhnicheskie nauki. 2012. № 3 (128). S. 37–41. (in Russian)
7. Pupkov K.A. Sovremennye metody, modeli i algoritmy intellektual'nykh sistem. Ucheb. posobie. M.: RUDN. 2008. (in Russian)