Agent Based Modeling and Deterministic Modeling of Eciton
Burchelli Swarm Patterns
06/07/11, 15:30 at Room 3631 (6th floor of building 3 of the Faculty of Sciences)
Eciton burchelli are a unique species of army ants that demonstrate
distinct swarm behavior. These ants have extremely poor eyesight, and
therefore, they depend on the distribution and tracking of pheromone
significantly in order to survive. A simulated ant world is created in this
project to observe the ants’ reactions to different initial food
distributions, and to determine the stability of their swarm patterns. The
goal of this project is to recover the swarm behavior observed in nature
using the simplest possible behavioral rules, and to investigate the impact
of intensity and direction of the pheromone trails on the emergent behavior
of ants through simulations. Several cases including a single food source,
multiple food sources, and perturbations to the multiple food sources
implemented by randomly placing barriers over established trails, are
studied in this project. In addition, a modified Fisher’s equation is
proposed to demonstrate ant swarm behavior. This equation has a traveling
wave solution that integrates the logistic growth and the diffusion of the
ants’ population density, and it incorporates the fact of pheromone trail
drift and loss of population during a swarm.
|
Back: Japanese / English |