Adaptive Multi-agent System: cooperation and structures emergence

Abstract

Nowadays, development of adaptive multi-agent systems has attracted recent attention. Adaptive systems are capable of adapting themselves to unforeseen changes of their environment in autonomous manner. We are interested, in this paper, to adaptation in cooperative systems where agents share a same goal. We propose an adaptive approach based on interaction evaluation and genetic algorithms. Agents, in this approach, have the ability to reorganize themselves in order to bring closer agents capable of cooperating in problem solving. The adaptation task is accomplished at both local and global levels. We present here details of this approach and we show some preliminary experimentation results.