Skip to main content
Fachgebiete
Automobil + Motoren Bauwesen + Immobilien Business IT + Informatik Elektrotechnik + Elektronik Energie + Nachhaltigkeit Finance + Banking Management + Führung Marketing + Vertrieb Maschinenbau + Werkstoffe Versicherung + Risiko
DE
EN
Themenseiten
Organisationspsychologie Projektmanagement Marketing Smart Manufacturing
Bücher
Zeitschriften
Weitere Formate
Podcasts Webinare Technik Webinare Wirtschaft Kongresse Veranstaltungskalender Awards
Jetzt Einzelzugang starten
Zugang für Unternehmen
Referenzkunden
MTZ-Fachkongress

Antriebe und Energiesysteme von morgen


Austausch von Automobil- und Energiewirtschaft

Am 14./15. Mai geht es in Chemnitz um die Mobilitätswende durch Elektro- und Wasserstofffahrzeuge.
Erweiterte Suche
Anmelden
nach oben

2015 | Buch

Introduction and Overview Kapitel lesen Erstes Kapitel lesen

From Pattern Formation to Material Computation

Multi-agent Modelling of Physarum Polycephalum

verfasst von: Jeff Jones

Verlag: Springer International Publishing

Buchreihe : Emergence, Complexity and Computation

Enthalten in: Springer Professional "Wirtschaft+Technik" , Springer Professional "Technik" , Springer Professional "Wirtschaft"

Einloggen, um Zugang zu erhalten
insite
SUCHEN

Über dieses Buch

This book addresses topics of mobile multi-agent systems, pattern formation, biological modelling, artificial life, unconventional computation, and robotics. The behaviour of a simple organism which is capable of remarkable biological and computational feats that seem to transcend its simple component parts is examined and modelled. In this book the following question is asked: How can something as simple as Physarum polycephalum - a giant amoeboid single-celled organism which does not possess any neural tissue, fixed skeleton or organised musculature - can approximate complex computational behaviour during its foraging, growth and adaptation of its amorphous body plan, and with such limited resources? To answer this question the same apparent limitations as faced by the organism are applied: using only simple components with local interactions. A synthesis approach is adopted and a mobile multi-agent system with very simple individual behaviours is employed. It is shown their interactions yield emergent behaviour showing complex self-organised pattern formation with material-like evolution. The presented model reproduces the biological behaviour of Physarum; the formation, growth and minimisation of transport networks.

In its conclusion the book moves beyond Physarum and provides results of scoping experiments approximating other complex systems using the multi-agent approach. The results of this book demonstrate the power and range of harnessing emergent phenomena arising in simple multi-agent systems for biological modelling, computation and soft-robotics applications. It methodically describes the necessary components and their interactions, showing how deceptively simple components can create powerful mechanisms, aided by abundant illustrations, supplementary recordings and interactive models. It will be of interest to those in biological sciences, physics, computer science and robotics who wish to understand how simple components can result in complex and useful behaviours and who wish explore the potential of guided pattern formation themselves.

Inhaltsverzeichnis

Frontmatter

Slime Mould Physarum Polycephalum

Frontmatter
Introduction and Overview
Abstract
This book concerns the computational approximation of the behaviour of the true slime mould Physarum Polycephalum, using a multi-agent approach. The giant single-celled organism has long been of biological interest, primarily due to its large size, complex internal streaming, and sensory-motor behaviour. In the last decade, however, it has been the subject of intense research into the complex computational properties it exhibits during its foraging, growth and adaptation.
Jeff Jones
Slime Mould Physarum Polycephalum
Abstract
This chapter introduces Physarum, describing the organism, its composition, habitat, behaviour and the biological basis of its network adaptation and oscillatory phenomena. We examine how Physarum approximates computational and robotic behaviour, relating it to classical and non-classical computing approaches. We explore previous modelling approaches to Physarum, framing its computational behaviour as a type of membrane-bound reaction diffusion computation. Finally, we set out a list of requisite behaviours which must be reproduced when modelling the organism.
Jeff Jones

Modelling Physarum Polycephalum

Frontmatter
A Multi-agent Model of Physarum
Abstract
This chapter describes the multi-agent model of Physarum, which is a particle based reaction-diffusion pattern mechanism behaving as a collective virtual material. The base model behaviour is described and its pattern formation properties explored in an evaluation of model parameters.
Jeff Jones
Modelling the Biological Behaviour of Physarum
Abstract
This chapter presents results reproducing a range of biological patterning observed in Physarum using the virtual plasmodium model. We demonstrate the initial formation of protoplasmic networks, and the growth and adaptation under differing nutrient concentration and substrate conditions. We conclude by suggesting how the model may provide clues as to the generation of apparently ‘intelligent’ behaviour of the plasmodium.
Jeff Jones
Implementing Neural Phenomena in Unorganised Non-neural Substrates
Abstract
Living organisms perceive their environment with a wide variety of special sensory modalities. Enhancing the contrast in the stream of information from these senses allows organisms to discriminate between small changes in signal level, potentially enhancing survivability. Lateral Inhibition (LI) is a neural mechanism which enhances the activity of neurons directly exposed to excitatory stimuli whilst suppressing the activity of their near neighbours (see Fig. 5.1 for a schematic illustration). LI phenomena have been described in auditory [173], somatosensory [174] and olfactory senses [175], but are most famously described in the visual systems of a wide range of animals, including humans [176, 177, 178].
Jeff Jones

Material Computation in a Multi-agent Model of Physarum Polycephalum: Mechanisms and Applications

Frontmatter
Modelling Computational Behaviour of Physarum
Abstract
In this part of the book we firstly examine how the multi-agent model reproduces the computational approximations first seen in slime mould which was initiated by the research of Nakagaki [183] who observed that the Physarum plasmodium was capable of solving simple maze problems. This initial work was significantly extended in terms of computational breadth, notably by the works of Adamatzky (see [4] for an overview). The first chapter in this section builds upon the simulation of the biological behaiour of Physarum given in Chapter 4 and examines how the model plasmodium computes by network formation and adaptation.
Jeff Jones
Approximating Classical Computing Devices with the Multi-agent Model
Abstract
In the previous chapter we examined how the network formation and adaptation of the model slime mould could be used to approximate computations. Although slime mould computes by means of its spatial propagation and shape adaptation, this is not the way the current dominant form of computation – classical computation – operates. In modern computer systems problems are abstracted into a symbolic representation and implemented by programs which are ultimately executed by microscopic operations involving transformation of input signals by logic gates.
Jeff Jones
Dynamical Reconfiguration of Transport Networks Using Feedback Control
Abstract
In previous chapters we have examined how the microscopic particle interactions of the multi-agent model of slime mould generate emergent behaviours which have material-like properties. Because this ‘virtual material’ is naturally adaptive in its shape, and because it can be influenced by external stimuli, it may potentially be classed as a so-called Smart Material. Smart materials are materials which can change their structural and/or functional properties in response to external stimuli [196].
Jeff Jones
Material Approximation of Combinatorial Optimisation
Abstract
The Travelling Salesman Problem (TSP) is a combinatorial optimisation problem well studied in computer science, operations research and mathematics. In the most famous variant of the problem a hypothetical salesman has to visit a number of cities, visiting each city only once, before ending the journey at the original starting city. The shortest path, or tour, of cities, amongst all possible tours is the solution to the problem. The problem is of particular interest since the number of candidate solutions increases greatly as n, the number of cities, increases.
Jeff Jones
Voronoi Diagrams and Their Variants with Attractant and Repulsion Fields
Abstract
The Voronoi diagram of a set of n points in the plane is the subdivision of the plane into n cells so that every location within each cell is closest to the generating point within that cell. Conversely the bisectors forming the diagram are equidistant from the points between them.
Jeff Jones
Material Representation of Area and Shape: Convex Hull, Concave Hull and Skeleton
Abstract
Computational geometry problems tackle the grouping or partitioning of points in the plane or in higher dimensions. Because of the lack of supportive tissue, the plasmodium typically extends along the space of the surface on which it lives, and hence, Physarum may be considered as a 2D organism.
Jeff Jones
Material Computation of Data Smoothing and Spline Curves
Abstract
Although Physarum slime mould has desirable computational properties, it also has some practical limitations. Although relatively simple and inexpensive to culture, its computation is slow, taking many hours – or even days – during which time it must be maintained within strict environmental parameters of temperature, light exposure and humidity. Physarum may also be relatively unpredictable in its behaviour which, although useful in wild conditions, is a hindrance when repeatability is concerned.
Jeff Jones
Tracking Statistical Properties and Changing Data via Morphological Adaptation
Abstract
The sclerotium stage is a part of the life cycle of Physarum, whose entry is provoked by adverse environmental conditions, particularly by a gradual reduction in humidity. In prolonged dry conditions the mass of plasmodium aggregates together, abandoning its protoplasmic tube network to form a compact, typically circular or elliptical, toughened mass [254]. Sclerotinisation protects the organism from environmental damage and the slime mould can survive for many months – or even years – in this dormant stage, reentering the plasmodium stage when moist conditions return. Biologically, the sclerotium stage may be interpreted as a primitive survival strategy and it has been interpreted computationally as a biological equivalent of freezing or halting a computation [96] in spatially represented biological computing schemes.
Jeff Jones
Morphological Adaptation Approaches to Path Planning
Abstract
Path planning (or motion planning) is a common application of computer science and robotics where a path has to be found between points (typically two points, source and destination point) within an arena. The representation of the arena may already be known or may be discovered by localisation and mapping methods (in this chapter we consider examples where the arena layout is known in advance). The resultant path should be short, minimising distance between the points. Other constraints may also apply, such as requiring paths of sufficient width, avoiding walls, avoiding obstacles, or minimising the number of turns.
Jeff Jones

From Emergent Oscillations to Collective Transport and Amoeboid Movement

Frontmatter
Emergence and Transitions of Spatio-temporal Oscillations
Abstract
In this chapter we experimentally investigate the re-generation and synchronisation of oscillation patterns in Physarum, and use the virtual plasmodium model to replicates the process. This chapter mainly consists of two parts: First we present experimental observation of oscillatory behaviour in the Physarum slime mould, and then present modelling results of oscillatory patterns, pattern transitions and synchronisation behaviour. Experimental work and data analysis using plasmodium of Physarum in sections 15.2 and 15.3 was performed by Dr. Soichiro Tsuda [259].
Jeff Jones
Modelling Collective Transport and Amoeboid Movement
Abstract
In this chapter we examine how the oscillatory phenomena described in the previous chapter may be patterned and harnessed to generate regular travelling waves which may be used to transport material within the virtual plasmodium, or to transport the virtual plasmodium itself by a cohesive amoeboid movement.
Jeff Jones

Conclusions and Beyond Physarum Models

Frontmatter
Summary of the Approach and Modelling beyond Physarum
Abstract
This book investigated the complex biological and computational behaviour of the true slime mould Physarum Polycephalum. The aim was to answer the question of how a simple single-celled organism, without any supposed requisites for intelligent behaviour, is capable of such complex feats of distributed and emergent computation.
Jeff Jones
Backmatter
Metadaten
Titel
From Pattern Formation to Material Computation
verfasst von
Jeff Jones
Copyright-Jahr
2015
Electronic ISBN
978-3-319-16823-4
Print ISBN
978-3-319-16822-7
DOI
https://doi.org/10.1007/978-3-319-16823-4

Premium Partner

    Bildnachweise