Meeting Abstract
When a European honey bee (Apis mellifera) colony outgrows its nest, the colony divides in two, sending a queen and about half the workers in search of a new home. To survive this transition, the outbound group will find a tree branch or other surface and hang together in a swarm while scouts search and decide on a permanent nest location. While this decision-making process of swarms has been well studied, the mechanical aspects of the formation of the swarm on its anchoring surface have not previously been studied. In this work, we will present both experimental and computational work on the formation of honey bee swarms to address the question: How do honey bees decide where to attach to a growing honey bee swarm? To gain insight into this question, we conduct experimental observations of swarm formation and replicate the observed behavior through computational modeling. With the queen confined to a cage, we induce the bees into a swarm under a horizontal surface and allow the bees to locate a new nest site and leave. Because the queen does not travel with them, we record the re-formation of the swarm as the bees return to her. Then, we use agent-based modeling techniques to model swarm formation as an aggregation problem, modeling the decision-making process of where bees join the swarm as it grows. With these modeling techniques, we are able to test various rules for swarm formation and assess their accuracy in reproducing our experimental results. A better understanding of the mechanics and behavior of honey bee swarms will allow for better design of, for example, self-assembling multi-agent robotic systems.
