Towards development in evolvable hardware

Abstract

Mapping between genotype and phenotype using a model of biological development has been widely touted as a technique for evolving solutions to large, complex problems [1-3]. Here we describe two test-bed developmental systems for evolvable hardware problems, and compare each to a naive mapping system. We find that designing evolvable developmental systems is not a trivial problem, however early analysis of the evolved structures demonstrates the potential of the generative processes behind development. We also account for the differences between the results of the two systems, highlighting the importance of search space evolvability over size.

Controlling Robots with Fractal Gene Regulatory Networks

“…Kumar, (with Wolpert) has developed extensive modelling software capable of evolving genes that are expressed into proteins, forming GRNs that control the expression of the genes and the development (growth, placement, function) of cells (Kumar and Bentley 2003c). Gordon has been investigating the use of development-inspired evolutionary systems for evolvable hardware, which makes use of ideas such as differentiation and growth (Gordon and Bentley 2002).…”

Ontogenetic Development and Fault Tolerance in the POEtic Tissue

“…The concept of protein gradients has been examined in detail in many publications, for biological [4,14] as well as electronic [5,6,9] organisms. Essentially, a gradient-based system consists of a set of diffusers that release a given protein into the system.…”

Biologically-Inspired: A Rule-Based Self-Reconfiguration of a Virtex Chip

“…Combining artificial development with evolution thus aims to improve the complexity of electronic circuits that may be evolved [3,4,5].…”