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
Bücher
Zeitschriften
Themenseiten
Organisationspsychologie Projektmanagement Marketing Smart Manufacturing
Weitere Formate
Podcasts Webinare Technik Webinare Wirtschaft Kongresse Veranstaltungskalender Awards
Jetzt Einzelzugang starten
Zugang für Unternehmen
Referenzkunden
SLX-Digitalkonferenz: Zukunftswerkstatt 2024

Mehr Umsatz mit Top-Kontakten

Am 7. Mai erfahren Sie, wie Vertriebsteams Leadgenerierung, Leadnurturing und Leadstrategien effizient steuern können.
Erweiterte Suche
Anmelden
nach oben

2016 | Buch

Introduction Kapitel lesen Erstes Kapitel lesen

Tracking and Mapping of Spatiotemporal Quantities Using Unicellular Swarm Intelligence

Visualisation of Invisible Hazardous Substances Using Unicellular Swarm Intelligence

verfasst von: John Oyekan

Verlag: Springer International Publishing

Buchreihe : Biosystems & Biorobotics

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

Einloggen, um Zugang zu erhalten
insite
SUCHEN

Über dieses Buch

The book discusses new algorithms capable of searching for, tracking, mapping and providing a visualization of invisible substances. It reports on the realization of a bacterium-inspired robotic controller that can be used by an agent to search for any environmental spatial function such as temperature or pollution. Using the parameters of a mathematical model, the book shows that it is possible to control the exploration, exploitation and sensitivity of the agent. This feature sets the work apart from the usual method of applying the bacterium behavior to robotic agents. The book also discusses how a computationally tractable multi-agent robotic controller was developed and used to track as well as provide a visual map of a spatio-temporal distribution of a substance. On the one hand, this book provides biologists and ecologists with a basis to perform simulations related to how individual organisms respond to spatio-temporal factors in their environment as well as predict and analyze the behavior of organisms at a population level. On the other hand, it offers robotic engineers practical and fresh insights into the development of computationally tractable algorithms for spatial exploratory and mapping robots. It also allows a more general audience to gain an understanding of the design of computational intelligence algorithms for autonomous physical systems.

Inhaltsverzeichnis

Frontmatter
Chapter 1. Introduction
Abstract
In the event of an invisible hazardous substance accidental leak or ecological disaster, avoiding the substance is a challenge to humans. This is because there is no visual reference to use in order to keep away from the polluted area or at least keep to the areas of lower substance concentration. Such a substance could include huge quantities of carbon dioxide as was the case in 1986 when the lake nyos in cameroon released an underwater storage of carbon dioxide into the air.
John Oluwagbemiga Oyekan
Chapter 2. Literature Review
Abstract
The necessity to monitor the environment is increasing everyday due to various issues related to environmental pollution. Pollution comes in various forms including gaseous, water, and even temperature pollution. Temperature pollution in the form of heat released from a nuclear plant’s exchanger, for example, makes an environment inhabitable for plankton and invariably affects the wild life that depend on the plankton to survive. This leads to a crash in the food chain that could lead to extinction of marine wild life populations.
John Oluwagbemiga Oyekan
Chapter 3. Investigative Process
Abstract
This chapter discusses the investigative process that was taken during the development of a biologically inspired coverage controller. The test case scenarios that were considered during development are also discussed in this chapter together with how the simulators used in this book were developed. In Chap. 1, it was mentioned that the goal was to be able to deploy a swarm of robots to form the distribution of an invisible dynamic spatiotemporal quantity. In addition, the swarm should also have the capability of responding to changes in the distribution of the quantity with the fluidity of a natural flock of starlings in flight.
John Oluwagbemiga Oyekan
Chapter 4. Developing and Implementing a Source Finding Controller
Abstract
In the literature review section of Chap. 2, it was mentioned that Berg and Brown derived a model of the bacteria chemotactic foraging behaviour based upon data obtained from experiments involving bacteria. The model obtained was a best fit curve on the experimental data. In this chapter, an investigation into how this model can be converted into a spatiotemporal source seeking controller for use on a robotic platform is conducted. As mechanisms used by biological organisms cannot sometimes be adopted directly, some modifications were introduced during the development of the model-derived controller in order to achieve this goal. The developed controller has parameters that make it possible to adjust the agent’s spatial exploitation process, environmental exploration and sensitivity to spatial readings.
John Oluwagbemiga Oyekan
Chapter 5. Relationship Between the Berg–Brown Model and the Keller–Segel Model
Abstract
In the previous chapter, it was observed that the agents using the Berg and Brown controller where able to follow and visually form the structure of the experimental plume used in the simulated environment. This behaviour was one that was emergent and was not programmed into the individual robots. In this chapter, investigation into this behaviour is conducted. It has been observed over the years that bacterial populations tend to form rings around food substrates when deployed in them. The type of ring formed varies from bacterial type to bacterial type.
John Oluwagbemiga Oyekan
Chapter 6. Behaviour Based Coverage Controller
Abstract
In this chapter, an attempt is made to solve the collision problem that was identified in the previous chapter. This was addressed by using the collision avoidance property of a flocking behaviour. In order to make use of the coverage properties of the bacteria behaviour with the collision avoidance property of a flocking behaviour, a behaviour based paradigm of robotics was used. This is unlike previous approaches such as the voronoi partition, deterministic annealing and virtual spring methods. It is also the first time this paradigm has been used in the context of providing visual representation of a spatiotemporal quantity. Furthermore, it is shown that machine learning could be used with the developed behaviour based architecture to fine tune the bacteria controller for optimal coverage.
John Oluwagbemiga Oyekan
Chapter 7. Improvements and Towards Real World Applications
Abstract
The natural world contains a lot of challenges that would make it difficult to deploy algorithms on physical agents. The challenges posed by the natural environment include noise both in sensor readings and vehicle dynamics and other unpredictable environmental conditions such as sudden wind changes, weather changes and so on. Unpredictability of environmental conditions often cause local maximums of the spatiotemporal quantity being monitored to develop in areas far away from the source. Organisms operating in the natural world have developed various mechanisms in order to deal with their unpredictable environment. For example, in order to deal with noise and the unpredictable dynamics in its natural environment, bacteria has developed a filtering mechanism.
John Oluwagbemiga Oyekan
Chapter 8. Conclusion
Abstract
In this book, the development of a controller that would enable the visual mapping or representation of an invisible spatiotemporal substance in the environment was conducted. The main requirements for the coverage controller was that it should be of minimal communication cost, computationally efficient and reactive. These requirements were chosen so that agents utilising the controller would be able to respond to dynamic changes in the distribution of the spatiotemporal substance with the fluidity of a flock of starlings in flight. If these requirements were met, then the developed controller would have an advantage over present coverage schemes that utilise machine learning and require high communication costs to achieve the same goal.
John Oluwagbemiga Oyekan
Metadaten
Titel
Tracking and Mapping of Spatiotemporal Quantities Using Unicellular Swarm Intelligence
verfasst von
John Oyekan
Copyright-Jahr
2016
Electronic ISBN
978-3-319-27425-6
Print ISBN
978-3-319-27423-2
DOI
https://doi.org/10.1007/978-3-319-27425-6

Neuer Inhalt

    Bildnachweise