Structurally Dynamic Cellular Automata
Andrew Ilachinski
Paul Halpern
Institute for Theoretical Physics, SUNY at Stony Brook,
Stony Brook, NY 11794, USA
Abstract
A new kind of cellular automaton (CA) model is introduced in which binary value-configurations and the (conventionally quiescent) underlying topological structure are dynamically coupled. Topology alterations are defined by local transition rules analogous to the value functions studied in conventional CA models defined on fixed lattices. Preliminary investigation reveals a rich spectrum of evolutionary behavior, including growth, decay, periodicity, and a relaxation to dynamic states with a stable effective dimensionality. The model offers a unique arena in which to study the emergence of true geometric self-organization and may provide the simplest approach to modeling the dynamics of the many different varieties of random cellular structures found in nature.
