elon}
Ryan T. Barnard (Professor Joel Hollingsworth), Department of Computing Sciences
Many of man's scientific advances have come from the
imitation of the natural world. With lessons learned from
nature, researchers are exploring new ideas in the fields of
Mathematics and Computer Science. Ant Colony
Optimization (ACO) is a relatively easy approach to finding
optimal solutions to difficult (NP-complete) problems. This
is done through the simulation of ant colonies, utilizing the
emergent intelligent behavior resulting from the interactions
of many ants. A single simulated ant shows very little
intelligent decision making skills, but a colony of simulated
ants have shown the ability to solve extremely complex
problems in a reasonable amount of time. This research,
Arbitrarily Extensible Ant Colony Optimization (AEACO),
extends ACO techniques to create a new algorithm that is
extensible, easily applied to new types of problems, and is
almost entirely context independent.
This talk will begin with an introduction to the
implementation of the discrete event simulator used to
simulate the ant colony. The extensions to ACO in order to
achieve AEACO will be developed. I will show how every
problem that can be defined as an n-dimensional adjacency
matrix can be solved using my techniques, specifically
focusing on food gathering as an example of the capabilities
of AEACO. Finally, I will discuss a comparison of various
approaches to solving the food gathering problem.