Dynamic control and information processing in the Belousov–Zhabotinsky reaction using a coevolutionary algorithm

Abstract

We propose that the behavior of nonlinear media can be controlled dynamically through coevolutionary systems. In this study, a light-sensitive subexcitable Belousov–Zhabotinsky reaction is controlled using a heterogeneous cellular automaton. A checkerboard image comprising of varying light intensity cells is projected onto the surface of a catalyst-loaded gel resulting in rich spatiotemporal chemical wave behavior. The coevolved cellular automaton is shown to be able to either increase or decrease chemical activity through dynamic control of the light intensity within each cell in both simulated and real chemical systems. The approach is then extended to construct a number of simple logical functions.

On dynamical genetic programming: simple Boolean networks in learning classifier systems

“…He shows an increase in performance over Mitchell et al's work by exploiting the potential for spatial heterogeneity in the tasks. More recently, a variation of this approach has been used to construct simple chemical logic gates within a non-linear reaction (e.g., [Toth et al, 2008]). Hence for twenty years it has been known that it is possible to design emergent computation from a graph-based discrete dynamical system through simulated evolution.…”

Are complex systems hard to evolve?

“…Any activity in automaton with rule AND became extinct after a very short transient period [ Figure 2 Hierarchies of evolutionary complexity, derived from studies on evolving logical gates in liquid crystal [4] and Belousov-Zhabotinsky medium [5], are consistent in the majority of cases studied. There is a direct correlation between the evolutionary hierarchy, based on minimal number of evaluations (Findings 1 and 2) and the behavioral complexity hierarchy (Finding 3).…”

“…We infer hierarchies of evolutionary complexity from computational and experimental laboratory studies of evolving of logic gates in liquid crystal [4] and excitable chemical medium [5]. See full details in the article cited, here we provide only a very brief overview of the results.…”

“…The characteristics of the evolutionary process for each of logical gates evolved are shown in Figure 1(a). These results by Harding and Miller [4] lead us to the following (''3'' means ''quicker to evolve than''): Toth et al [5] evolved Boolean gates AND, NAND, and XOR in a hybrid system ''cellular automaton-excitable chemical medium.'' A light-sensitive modification of the Belousov-Zhabotinsky reaction was employed.…”

“…Nonlinear spatial extended media-liquid crystal [4] and Belousov-Zhabotinsky system [5]-were chosen as substrates of various logical gates' evolution because these media are possibly the best prototypes of nonclassical, unconventional, computing devices [6]. One-dimensional cellular automata were then selected as substrates to implement the gates evolved because the automata are good approximators of the real world [7], the space-time dynamic of the automata is easy to visualize, and methods for exploring complexity of the automata are well-established [8][9][10][11][12].…”