Partner und Internationale Organisationen
(Englisch)
|
University of York (UK), EPFL (ASL-STI and LSL-I&C), Lausanne (CH), UNIL, Lausanne (CH), University of Glasgow (UK); Technical University of Catalunya (UPC), Barcelona (E)
|
Abstract
(Englisch)
|
The goal of this project is the development of a novel digital electronic circuit, a flexible computational substrate or artificial tissue, capable of integrating the three biological models of self-organization: phylogeny (P), ontogeny (O), and epigenesis (E). This tissue will be the essential substrate for the creation of POE-based machines, capable of evolution, growth, self-repair, self-replication, and learning. The POEtic tissue will be a cellular surface composed of a variable number of elements, or cells. Each cell will contain the entire description (genome) for the whole tissue and will have the ability to communicate with the environment (through sensors and actuators) and with neighboring cells and accordingly executing a function. During the project, the tissue will be validated on different kinds of applications that can benefit from its life-like properties, interacting with and adapting to their own environment.
Objectives:
(1) To review evolutionary models applied to electronic circuits.
(2) To develop suitable genetic encodings and analyze their effects on fitness landscape.
(3) To develop a set of hardware mechanisms, inspired upon embryology, that will provide growth, self-repair, and self-replication of the electronic tissue.
(4) To review neuronal models and architectures suitable for digital implementation.
(5) To define and design O, PO, PE, OE, and POE cellular tissues including the properties defined earlier.
(6) To demonstrate the applicability of these tissues on a variety of applications where its effectiveness can be readily assessed qualitatively and quantitatively.
Note: P, O and E relate to the three biological models: phylogenesis (P), ontogenesis (O), and epigenesis (E).
Work description:
The POEtic tissue will be a cellular surface composed of a variable number of elements, or cells. Each cell will have the ability to communicate with the environment (through sensors and actuators) and with neighbouring cells (through bi-directional channels), and accordingly executing a function. Each cell of the tissue will have the same basic structure, but will be able to acquire different functionalities, as 'totipotent cells' in living organisms. This flexibility will be given by an organization in three layers: a genotype plane, a configuration plane, and a phenotype plane. The genotype plane of each cell will contain a full description of the organism in the form of a digital genome. The configuration plane will transform the genome into a configuration string directly controlling the processing unit of the phenotype plane. Through this cellular process, the tissue will be organized into a massively parallel multi-cellular electronic structure. Within such structure, groups of cells will be able to co-operate to realize a given task, giving rise to substructures not unlike organs in living beings. It is important to notice that the tissue can implement the three models (P, O, and E) in any possible combination, providing a very powerful computational substrate. Indeed, we shall proceed with a gradual approach. We shall start by implementing an ontogenetic tissue (O), then separately implement the three possible combinations of two models (PO, PE, and OE), and finally we shall realize a tissue that integrates all three models (POE). Such POEtic tissue will represent our final living artefact. During the whole duration of the project, the five tissues will be validated on different kinds of applications that can benefit from the life-like properties of our artefacts. The results of these applications are not to be considered as a main goal of this project, but only as prototypes to show different aspects of the tissue.
Note: P, O and E relate to the three biological models: phylogenesis (P), ontogenesis (O), and epigenesis (E).
|
