Towards Artificial Evolution of Complex Behaviors Observed in Insect Colonies

Abstract

AbstractStudies on social insects have demonstrated that complex, adaptive and self-organized behavior can arise at the macroscopic level from relatively simple rules at the microscopic level. Several past studies in robotics and artificial life have focused on the evolution and understanding of the rules that give rise to a specific macroscopic behavior such as task allocation, communication or synchronization. In this study, we demonstrate how colonies of embodied agents can be evolved to display multiple complex macroscopic behaviors at the same time. In our evolutionary model, we incorporate key features present in many natural systems, namely energy consumption, birth, death and a long evaluation time. We use a generic foraging scenario in which agents spend energy while they move and they must periodically recharge in the nest to avoid death. New robots are added (born) when the colony has foraged a sufficient number of preys. We perform an analysis of the evolved behaviors and demonstrate that several colonies display multiple complex and scalable macroscopic behaviors.KeywordsSocial InsectTask AllocationIntermediate ZoneMacroscopic BehaviorInsect ColoniThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.