A Mobile Agent-based Coalition Formation System for Multi-robot Systems

This paper presents a bio-inspired mobile agent-based coalition formation system for recruiting modular robots into different teams. A mathematical model for the coalition formation problem is described in this article. A mobile agent-based system is presented to solve the problem described. The system aims at bridging the gap between theory and practice in robots coalition formation. With this system, it is able to recruit a large scale of robots into many different groups effectively and efficiently for a variety of applications, such as parallel performance of multiple tasks by multiple teams of robots. Unlike the centralized system, each robot will make its own decision based on the ant-colony algorithm to guarantee the global coalition formation efficiency. Also, due to the benefits of the mobile agent technology, the system has a high flexibility to integrate and adopt various algorithms to meet different requirements of robot recruitment. An overall system architecture and implementation details are presented in the article, as well as a coalition formation example to show the properties of the system.