Bio-inspired self-organization for supporting dynamic reconfiguration of modular agents

This paper presents an approach to develop and manage a self-organizing, distributed computing system in dynamically changing environments. In the proposed approach, a target application is characterized by a collection of distributed, autonomous, and diverse mobile agents, designed to mimic the adaptive behavior of biological systems. Each mobile agent is implemented as dynamic modular components, which can automatically migrate and reconfigure while the application is being executed. Assemblage of modular components can be altered according to the local migration schemes of agents, which include deployment behavior based on biological processes. This paper presents an adaptive architecture, which can reorganize and reconfigure agents based on several different biological concepts and mechanisms. We describe several key features of the modular components, the design methodologies and implementation of the modular, self-organizing framework, and demonstrate how the framework satisfies a set of functional requirements derived from specific agent behavior. We also present some simulation results to examine the scalability and efficiency of the framework.