Study of Ant Colony Optimization Technique for Coalition Formation in Multi Agent Systems

The field of artificial intelligence has given rise to the concept of multi agent systems. An agent in computer science terms is a computer program which acts on behalf of a user or another program to accomplish a predefined task. Coalition formations amongst agents represent a kind of grouping in which they cooperate with each other to effectively achieve their own goal or a common goal. Based on the final goal achievement, agents can be grouped into two categories. First are Individual, autonomous agents which focuses on fulfilling their own goal and are hence termed as selfish agents. The other kind of agents is cooperative agents which put the entire system goal before their own goal. The later kind of agents forms the core part of this paper. Agents in a multi agent system may need to cooperate in order to complete the tasks. We form a group of agents from the set of agents in cases where a single agent may not complete the given task or if it complete, the performance is not optimal from the systems point of view. This paper presents the ant colony optimization technique to coalition formation problem in MAS.