Evolving Self-organizing Cellular Automata Based on Neural Network Genotypes

Abstract

This paper depicts and evaluates an evolutionary design process for generating a complex self-organizing multicellular system based on Cellular Automata (CA). We extend the model of CA with a neural network that controls the cell behavior according to its internal state. The model is used to evolve an Artificial Neural Network controlling the cell behavior in a way a previously defined reference pattern emerges by interaction of the cells. Generating simple regular structures such as flags can be learned relatively easy, but for complex patterns such as for example paintings or photographs the output is only a rough approximation of the overall mean color scheme. The application of a genotypical template for all cells in the automaton greatly reduces the search space for the evolutionary algorithm, which makes the presented morphogenetic approach a promising and innovative method for overcoming the complexity limits of evolutionary design approaches.

Engineered self-organization for resilient robot self-assembly with minimal surprise

“…Except for Conway's Game of Life [38], all of the presented approaches aim for the creation of predefined patterns either specifying the target fully [40,45] or partially using local templates [37]. Fitness is given for high resemblance to these structures.…”

Comparison of Evolutionary Development of Cellular Automata Using Various Representations

“…Other soft-computing techniques have also been investigated in relation with CA. For example, Elmenreich et al proposed an technique for calculating the transition function of CA using neural networks [15]. The goal was to train the network by means of Evolutionary Programming [16] in order to develop self-organising structures.…”

Atmospheric dispersion modeling using Artificial Neural Network based cellular automata

“…Depending on the aim of the study, different neighborhoods can be set such as Moore Learning: as Artificial Neural Networks (Almeida et al, 2008), genetic algorithms (Ak et al, 2013), self-organizing systems (Elmenreich and Fehérvári, 2011), Markov chain (Balzter et al, 1998), Monte Carlo simulations (Zio et al, 2006), fuzzy logic (Wu, 1998) unfortunately, no comparison was made with respect to actual tests or numerical simulations.…”