Ian-Based Cooperative Control Model for Multi-agent System

This study presents an Immune Agent Networks (IAN) control model. Based on this model, a behavioral control paradigm is developed. Effective coordination and mutual understanding between agents can be achieved by adopting such a behavioral control based on their corresponding behavior. Each agent is abstracted as an independent entity that searches for environment and exhibits robust behavior to accomplish tasks. In this article, simulations are presented with a box-collecting system. The significance of the behavioral control paradigm and the impact of the IAN-based behaviors on the overall performance of the box-collecting system are examined. We furthermore try to evolve affinities among antibodies using testing method. The simulation results validate the efficiency of the proposed IAN model and the importance of paratope mapping evolving is pointed out and related optimization methods that can be used to evolving the paratope mapping are illustrated.