Design Through Evolutionary Swarm Systems

By:
To add a paper, Login.

The performance level of current computer technologies opens the door for applications of evolutionary computing to serve as a symbiotic design partner for artists and designers. Creation, breeding, growing and evolving forms emulating the construction processes of Nature allow the designer to craft and mimic form through an understanding of the infrastructure provided by Nature's rudimentary building processes. This is similar to the way termites and ants build mounds or anthills. Each individual termite does not understand its unique role in the design of a mound; however, as it relentlessly performs several functions in the total community, a habitat is established. Evolutionary design systems also start with minuscule cell division, building upon this foundation to the scope of mega-engineering projects defined through complicated materials. The elements defining a dynamic system are “swarms” (virtual building agents that posses simplistic intelligence, yet when combined into colonies, create a communal intelligence capable of designing and building complexity). The swarm-based evolutionary design system starts with spartan simplicity. However, as the colony grows, the level of complexity is increased until the design becomes an evolutionary system capable of building and editing itself through the strategic definition of algorithmically defined fitness functions. Instead of the artist imitating Nature through landscape painting or derivative sculptures, Evolutionary Swarm Systems actually create and design (almost) like Nature. What Nature has defined serves as inspiring departure points from which to expand design “beyond” Nature’s offerings. We propose to present our latest work that extends on SwarmArt, Swarm Grammars and Swarm Architectures, which are computer implementations of evolutionary swarm-based building processes that are modelled in virtual 3-dimensional space.


Keywords: Evolutionary Design, Nature-Inspired Design, Swarm Intelligence, Swarmart, Swarm Design
Stream: Other
Presentation Type: Paper Presentation in English
Paper: A paper has not yet been submitted.


Dr. Christian Jacob

Associate Professor, Computer Science, University of Calgary
Calgary, Alberta, Canada

I received my Ph.D. in Computer Science from the University of Erlangen-Nuremberg, Germany. In July 1999, I joined the Department of Computer Science (Faculty of Science) at the University of Calgary. Since August 2003, I also hold a joint appointment with the Department of Biochemistry & Molecular Biology (Faculty of Medicine), where I am the Director of Bioinformatics in the Bachelor of Health Sciences Program. I lead the Evolutionary & Swarm Design (E&SD) research group of the Artificial Intelligence Laboratory. My research group is investigating how to apply evolutionary, swarm and collective intelligence techniques in various application domains. So far, the E&SD research group has built mathematical models, computer simulations and visualizations of traffic systems, army ants, neuron growth, biomolecular systems, gene regulatory systems, and swarm architectures. Some of the projects are described in detail at the ESD website: http://www.swarm-design.org. I have written two books on evolutionary computing and natural programming paradigms: Principia Evolvica (dpunkt, Heidelberg, 1997; in German) and Illustrating Evolutionary Computation with Mathematica (Morgan Kaufmann, San Francisco, 2001).

Prof. Gerald Hushlak

Professor, Art, University of Calgary
Calgary, Alberta, Canada


Ref: A09P0343