Preview

Mekhatronika, Avtomatizatsiya, Upravlenie

Advanced search

Method of Decentralized Control of the Distributed System during Execution of the Task Flow

Abstract

This paper is devoted to the distributed control system with the network architecture consisting of a multitude of objects united by a communication channel and participating in implementation of the flow of the incoming consumer tasks. At that, it is assumed that each consumer task consists of a set of interconnected operations, presented as an acyclic graph, and can appear any moment. In the paper the authors show that a centralized management solution with a single control unit in such a distributed system with a large number of objects is very complicated. Therefore, they propose a method for a decentralized management of the distributed systems using multiple management devices for the individual objects. They propose new algorithms for individual management of the objects of a distributed system for four versions of the original statement: the objects can perform the same set of operations at the same time; the objects perform the same set of operations, but the times of execution of the identical operations in various objects are different; the objects perform different sets of operations, but the times of execution of the identical operations are equal; the objects perform different sets of operations and the times of execution of the identical operations in different objects are different. In conclusion the authors present the results of the experimental researches of the proposed algorithms achieved due to application of the program model of the distributed system.

About the Authors

A. I. Kalyaev
Scientific Research Institute of Multiprocessor Computer Systems named after Academician A. V. Kalyaev
Russian Federation


I. A. Kalyaev
Scientific Research Institute of Multiprocessor Computer Systems named after Academician A. V. Kalyaev
Russian Federation


References

1. Юревич Е. И. О проблеме группового управления роботами // Мехатроника, автоматизация, управление. 2004. № 2. С. 9-13.

2. Каляев И. А., Мельник Э. В. Децентрализованные системы компьютерного управления. Ростов-на-Дону: ЮНЦ РАН, 2011. 196 с.

3. Интеллектуальные роботы / Под редакцией Е. И. Юревича. М.: Машиностроение, 2007. 360 с.

4. Каляев И. А. Метод коллективного управления группой объектов // Мехатроника, автоматизация, управление. 2004. № 3. С. 9-15.

5. Каляев И. А., Гайдук А. P., Капустян С. Г. Модели и алгоритмы коллективного управления в группах роботов. М.: Физматлит, 2009. 278 с.

6. Зуховицкий С. И., Радчик И. А. Математические методы сетевого планирования. М.: Наука, 1965. 360 с.

7. Кофман А., Дебезей Г. Сетевые методы планирования. М.: Прогресс, 1968. 182 с.

8. Новиков Д. А. Управление проектами. Организационные механизмы. М.: ПМСОФТ, 2007. 140 с.

9. Рейнгольд Э., Нивергельт Ю., Део Н. Комбинаторные алгоритмы. Теория и практика. М.: Мир, 1980. 476 с.


Review

For citations:


Kalyaev A.I., Kalyaev I.A. Method of Decentralized Control of the Distributed System during Execution of the Task Flow. Mekhatronika, Avtomatizatsiya, Upravlenie. 2015;16(9):585-598. (In Russ.)

Views: 480


Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.


ISSN 1684-6427 (Print)
ISSN 2619-1253 (Online)