TU Darmstadt / ULB / TUbiblio

Simulation of Autonomous Robot Teams with Adaptable Levels of Abstraction

Friedmann, Martin (2010):
Simulation of Autonomous Robot Teams with Adaptable Levels of Abstraction.
TU Darmstadt, [Online-Edition: urn:nbn:de:tuda-tuprints-21132],
[Ph.D. Thesis]

Abstract

Simulation is one of the most important Tools for the development of soft- and hardware for autonomous robots. Depending on the purpose of the simulation different levels of abstraction are required for the simulation of a robot's motion and sensors as well as its interaction with the environment. The present thesis discusses the simulation of autonomous robots with different levels of abstraction. Methods are derived which allow the simultaneous simulation of multiple robots within the same simulation using different levels of abstraction, thus allowing to use the simulation for differing purposes. Besides methods for the simulation of a robot's motion and sensors on different levels of abstraction, methods for adaptable modeling of the robots and validation of the simulation are discussed. The developed methodology is the base for the "Multi-Robot-Simulation-Framework". This framework is the foundation for several simulations of wheeled, humanoid and four-legged robots which are described as examples in this thesis.

Item Type: Ph.D. Thesis
Erschienen: 2010
Creators: Friedmann, Martin
Title: Simulation of Autonomous Robot Teams with Adaptable Levels of Abstraction
Language: English
Abstract:

Simulation is one of the most important Tools for the development of soft- and hardware for autonomous robots. Depending on the purpose of the simulation different levels of abstraction are required for the simulation of a robot's motion and sensors as well as its interaction with the environment. The present thesis discusses the simulation of autonomous robots with different levels of abstraction. Methods are derived which allow the simultaneous simulation of multiple robots within the same simulation using different levels of abstraction, thus allowing to use the simulation for differing purposes. Besides methods for the simulation of a robot's motion and sensors on different levels of abstraction, methods for adaptable modeling of the robots and validation of the simulation are discussed. The developed methodology is the base for the "Multi-Robot-Simulation-Framework". This framework is the foundation for several simulations of wheeled, humanoid and four-legged robots which are described as examples in this thesis.

Divisions: 20 Department of Computer Science
20 Department of Computer Science > Simulation, Systems Optimization and Robotics Group
Zentrale Einrichtungen
Exzellenzinitiative
Exzellenzinitiative > Graduate Schools > Graduate School of Computational Engineering (CE)
Exzellenzinitiative > Graduate Schools
Date Deposited: 12 May 2010 10:22
Official URL: urn:nbn:de:tuda-tuprints-21132
License: only the rights of use according to UrhG
Referees: von Stryk, Prof. Dr. Oskar and Pagello, Prof. Dr. Enrico
Refereed / Verteidigung / mdl. Prüfung: 30 November 2009
Alternative Abstract:
Alternative abstract Language
Eines der wichtigsten Werkzeuge bei der Entwicklung von Soft- und Hardware für autonome Roboter ist deren Simulation. Abhängig vom Zweck der Simulation sind dabei unterschiedliche Abstraktionsgrade bei der Simulation der Aktorik und Sensorik wie auch bei der Betrachtung der Interaktion mit der Umgebung erforderlich. Die vorliegende Dissertation behandelt die Simulation autonomer Roboter unter Verwendung verschiedener Abstraktiongrade. Dazu werden Methoden entwickelt, die es gestatten, mehrere Roboter mit verschiedenen Abstraktionsgraden innerhalb einer Simulation zeitgleich zu simulieren. Dies ermöglicht es, die selbe Simulation für unterschiedliche Zwecke einzusetzen. Neben Simulationsverfahren für die Aktorik und Sensorik von Robotern mit unterschiedlichen Abstraktionsgraden, werden in der vorliegenden Arbeit die Themen anpassbare Modellierung der Roboter und Validierung der Simulation behandelt. Basierend auf den entwickelten Methoden wurde das "Multi-Robot-Simulation-Framework" entwickelt. Dieses bildet die Grundlage für verschiedene Simulation radgetriebener, humanoider und vierbeiniger Roboter, die als Anwendungsbeispiele in der vorliegenden Arbeit beschrieben werden.German
Export:
Suche nach Titel in: TUfind oder in Google

Optionen (nur für Redakteure)

View Item View Item