S.А. Karpov - Post-graduate Student, Department «Management problems», Moscow State Institute of Radio Engineering, Electronics and Automation (MSTU MIREA). E-mail: s_karpov@mirea.ru P.E. Tripolski - Ph.D. (Eng.), Associate Professor, Department «Management problems», Moscow State Institute of Radio Engineering, Electronics and Automation (MSTU MIREA). E-mail: tripolski@mirea.ru

To complete tasks it is necessary to use multiple autonomous robots, combining them in multi-robot system. In this case, one of the most important problems is to create troubleproof failover communication between agents. This paper discusses ways to organize a unified network between all agents of multi-robot system using the most rational way. The analysis of the possible solutions to the problem has been done in this paper. The possible of implementation of the algorithm deter-mining the connectivity of robots of the multi-agent robotic system and possible ways of combining agents in a network are given. This paper contains the algorithms for finding minimal paths in graphs and method of combining vertices of the graph via network of the minimum total path the edges; the following problems are solved: calculating the number of agents required for the organization of a unified wireless communication in multi-agent robotic system; identify ways to create the wireless network for all the agents of the multi-robot system; calculation of coordinates of robots that provide continuous communication in multi-agent systems. The resulting algorithms form the basis of software that allows you to simulate the behavior of multi-agent robotic system with a given number of robots to evaluate the link at any time of the task, and propose a way to combine unrelated agents using a minimum number of additional robots. The results of this research can be used in estimation the effectiveness of the number of groups of agents in multi-agent robotic system, as well as in the planning of movements and in the allocation of tasks.

 

1. Multi-sensor ATTenuation Estimation (MATTE): Signal-strength prediction for teams of robots, Johannes Strom and Edwin Olson. U.S. DoD Grant FA2386-11-1-4024. Department of Computer Science and Engineering. UniversityofMichigan. AnnArbor. MI 48109 USA.
2. Zadacha SHtejjnera. Rezhim dostupa: http://ru.wikipedia.org/wiki/Zadacha_SHtejjnera.
3. Ivanov A.O., Tuzhilin A.A. Teorija ehkstremalnykh setejj. Moskva-Izhevsk: Institut kompjuternykh issledovanijj. 2003.
4. Algoritmy o vybore dorogi i setjakh. Seti SHtejjnera. Lekcija Vladimira Protasova. Rezhim dostupa: http://habrahabr.ru/company/yandex/blog/215931.
5. Ivanov A.O., Tuzhilin A.A. Zadacha SHtejjnera na ploskosti ili ploskie minimalnye seti // Matematicheskijj sbornik. 1991. T. 182. № 12. S. 1813-1844.