Universality of Evolved Cellular Automata in-Materio

Abstract

Evolution-in-Materio (EIM) is a method of using artificial evolution to exploit physical properties of materials for computation. It has previously been successfully used to evolve a multitude of different computational devices implemented in physical materials. One of the biggest problems in exploiting materials is finding a good computational abstraction to carry computation on top of the underlying physical process. This paper presents elementary cellular automata (CA) as a possible abstraction and presents successfully evolved CA transition functions in single-walled carbon nanotube (SWCNT) and polymer composite materials. Such implementation allows reasoning about the computational capabilities of materials and draw analogies with cellular automata complexity and computation at the Edge of Chaos. This work is done within the European Project NASCENCE.