TU Darmstadt / ULB / TUbiblio

Grid-Based Obstacle Mapping and Gaze Control for Soccer Playing Humanoid Robots

Stumpf, Alexander (2010):
Grid-Based Obstacle Mapping and Gaze Control for Soccer Playing Humanoid Robots.
Darmstadt, Technische Universität Darmstadt, Department of Computer Science (SIM), [Bachelor Thesis]

Abstract

An important research topic at the Simulation, Systems Optimization and Robotics group at TU Darmstadt is research and development of autonomous robot systems. One of the research projects is the Darmstadt Dribblers, developing autonomous soccer playing humanoid robots and participating in the annual RoboCup competition. In this work a new approach for modeling the environment of the robot is introduced. For this purpose an occupancy grid mapping system is developed which provides a more detailed model of obstacles in the dynamic environment than previously used approaches. Furthermore, an entropy-based active gaze control system generating gaze commands to maximize perceptual input is introduced. This system accounts for the reliability estimates in existing modules like ball modeling and is used to automatically control the camera gaze of the robot. Evaluation of the system shows that the system performs well on real hardware.

Item Type: Bachelor Thesis
Erschienen: 2010
Creators: Stumpf, Alexander
Title: Grid-Based Obstacle Mapping and Gaze Control for Soccer Playing Humanoid Robots
Language: English
Abstract:

An important research topic at the Simulation, Systems Optimization and Robotics group at TU Darmstadt is research and development of autonomous robot systems. One of the research projects is the Darmstadt Dribblers, developing autonomous soccer playing humanoid robots and participating in the annual RoboCup competition. In this work a new approach for modeling the environment of the robot is introduced. For this purpose an occupancy grid mapping system is developed which provides a more detailed model of obstacles in the dynamic environment than previously used approaches. Furthermore, an entropy-based active gaze control system generating gaze commands to maximize perceptual input is introduced. This system accounts for the reliability estimates in existing modules like ball modeling and is used to automatically control the camera gaze of the robot. Evaluation of the system shows that the system performs well on real hardware.

Place of Publication: Darmstadt
Divisions: 20 Department of Computer Science
20 Department of Computer Science > Simulation, Systems Optimization and Robotics Group
Date Deposited: 09 Jul 2019 10:44
Related URLs:
Export:
Suche nach Titel in: TUfind oder in Google

Optionen (nur für Redakteure)

View Item View Item