The field of multi-agent systems is concerned with societies of autonomous agents that interact to efficiently achieve their goals. In this work, we evolve in silico teams of agents that are capable of displaying division of labor. We hope to broaden the understanding of the evolutionary dynamics of fixed and adaptive mechanisms of specialization (division of labor). The purpose of our research is twofold. From engineering perspective, we wish to improve techniques of team optimization and intend to explorareae the of bio-inspired task allocation algorithms. From biological perspective, we hope to shed some light on evolutionary history and mechanical explanations of division of labor in social insects.
Research in the field of multi-agent systems spans both natural and engineering sciences (Fig. 1). Engineers uncover the complexities arising from the interactions between multiple agents, seeking for distributed control algorithms that would result in the desired joint behavior of the entire system. Biologists study insect societies, seeking for evolutionary roots and mechanistic explanations for behavioral traits observed in colonies of ants, bees and termites.
Figure 1. Examples of multi-agent systems
Ants nest in the forest of eastern North America (c) Alex Wild, www.myrmecos.net
Termite soldiers rushing forward to guard a breach in their next (c) Alex Wild, www.myrmecos.net
Micro-robots gathering pucks (c) lis.epfl.ch
Team of robots exploring the area (c) Jiuguang Wang, Wikimedia Commons
Cellular automata implementing Conway’s game of life (c) Justin Windle, blog.soulwire.co.uk
Computational cluster balancing the load (c) lacal.epfl.ch