The Swarm Application Framework

A Swarm Framework

Author: Don Miner
Advisor: Marie desJardins

The goal of this research is to create a framework for swarm applications. Application designers will be able to describe and command their swarms at the swarm level, opposed to manipulating basic rules for agents. This is all made possible with our method of "Rule Abstraction." For example, instead of telling a group of boids to avoid closeness, match velocity, and tend to the center of mass, the application designer can simply tell the group of agents to "flock." For a more information, consult some of the writings that have been done for this project, listed below.

Goals

Make the development of swarm applications intuitive.
Make rule abstraction a simple process, thus promoting the use of rule hierarchies.
Provide a collection of modules that developers can use to quickly build new applications.
Enable engineers to focus on the behaviors and properties of the swarm, rather than on low-level details of individual agent behaviors.

Publications and Documents

``Learning Abstract Rules for Swarm Systems" by Don Miner, Peter Hamilton, and Marie desJardins. Not published.
A paper that presents the basics of rule abstraction. This paper serves as a proof of concept.
``The Swarm Application Framework" by Don Miner, Marie desJardins, and Peter Hamilton. AAAI '08 Student Abstract.
A short (2 pages) introduction on SAF, which discusses motivation and architecture.

Presentations

``Abstract Rules for Swarm Systems" 4/25/08
Gives an short overview of rule abstraction, abstrat rule learning, and the form circle experiment.
``The Swarm Application Framework, Rule Abstraction and Future Directions" 3/5/08
Briefly talks about the fundamentals of SAF and Rule Abstraction and then provides a discussion of possible future directions.
``Swarm Intelligence and The Swarm Application Framework" 2/25/08
Gives an introduction to Swarm Intelligence and gives an overview of SAF. A talk given to Sue Evans' freshman seminar class at UMBC.

Other Output

Some interesting videos that pertain to this project.

This material is based upon work supported by the National Science Foundation under Grant No. 0545726.
Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation.

Back