Authors:
Ivan Babanin
1
;
Ivan Pustovoj
1
;
Elena Kleimenova
2
;
Sergey Kozhevnikov
1
;
Elena Simonova
1
;
Petr Skobelev
1
and
Alexander Tsarev
1
Affiliations:
1
Software Engineering Company «Smart Solutions» and Ltd., Russian Federation
;
2
Institution of the Russian Academy of Sciences Institute for the Control of Complex Systems of RAS, Russian Federation
Keyword(s):
8 Queens Problem, Evolutionary Computing, Multi-agent Technology, Strategy of Conflict’s Resolving, Domain Ontologies, Experimental Data.
Related
Ontology
Subjects/Areas/Topics:
Artificial Intelligence
;
Computational Intelligence
;
Evolutionary Computing
;
Evolutionary Robotics and Intelligent Agents
;
Soft Computing
;
Swarm/Collective Intelligence
Abstract:
The problem of 8 Queens is one of the most well-known combinatorial problems. In this article multi-agent evolutionary-based solution for ‘8 Queens’ problem is proposed. In the multi-agent solution each Queen (or other chess-man) gets a software agent that uses a 'trial-and-error' method in asynchronous and parallel decision making on selecting new position for queens. As the result the solution is found in distributed manner without main control center that provides a number of benefits, for example, introducing new types of chess-man or changing constraints in real time. Two main strategies of Queen’s decision making process has been considered and compared in experiments: random generation of the next move and conflict-solving negotiations between the agents. Experiments’ results show significant acceleration of the decision making process in case of negotiation-based strategy. This solution was developed for training course for students of Computer Science as a methodical basis f
or designing swarm-based multi-agent systems for solving such complex problems as resource allocation and scheduling, pattern recognition or text understanding.
(More)