A Fuzzy Logic-Based Approach for Flexible Self-Interested Agent Team Forming

In our research, we try to develop a mechanism that can enable selfinterested agents to flexibly choose team durations and objectives. Factors, such as agent historical performances, task requirements and resource constraints, should be considered in the mechanism. Especially for open environments, the flexible team forming mechanism will be more suitable for self-interested agent applications. It enables more dynamic and reasonable collaboration between agents and reduce unnecessary consumptions brought by team forming. However, due to the highly uncertainties of most open environments, analysis and evaluations of dynamic factors is not very easy. Especially, it is impossible to find a fixed standard for factors such as “how good an agent performance is.” Regarding this point, in this paper, we present a fuzzy logic based team forming approach that can enable agents perform those evaluation according to some fuzzy rules. In the approach, an agent can dynamically select collaboration durations and objectives according to result of fuzzy evaluations, and choose collaboration manners more flexibly.

The rest of this paper is arranged as follows. In the second section, we introduce the MAS structures and some important definitions and assumptions that are related with the approach. In Sect. 3, we analyze the advantages, disadvantages and suitable areas of long-term and one-shot teams. The fuzzy logic based approach is introduced in Sect. 4. Finally, the conclusions and further directions of this research are presented in Sect. 5.