Item


Intelligent task-level grasp mapping for robot control

In the future, robots will enter our everyday lives to help us with various tasks. For a complete integration and cooperation with humans, these robots need to be able to acquire new skills. Sensor capabilities for navigation in real human environments and intelligent interaction with humans are some of the key challenges. Learning by demonstration systems focus on the problem of human robot interaction, and let the human teach the robot by demonstrating the task using his own hands. In this thesis, we present a solution to a subproblem within the learning by demonstration field, namely human-robot grasp mapping. Robot grasping of objects in a home or office environment is challenging problem. Programming by demonstration systems, can give important skills for aiding the robot in the grasping task. The thesis presents two techniques for human-robot grasp mapping, direct robot imitation from human demonstrator and intelligent grasp imitation. In intelligent grasp mapping, the robot takes the size and shape of the object into consideration, while for direct mapping, only the pose of the human hand is available. These are evaluated in a simulated environment on several robot platforms. The results show that knowing the object shape and size for a grasping task improves the robot precision and performance

Manager: García Campos, Rafael
Other contributions: Universitat de Girona. Escola Politècnica Superior
Author: Comas Jordà, Josep Maria
Date: 2006 June
Abstract: In the future, robots will enter our everyday lives to help us with various tasks. For a complete integration and cooperation with humans, these robots need to be able to acquire new skills. Sensor capabilities for navigation in real human environments and intelligent interaction with humans are some of the key challenges. Learning by demonstration systems focus on the problem of human robot interaction, and let the human teach the robot by demonstrating the task using his own hands. In this thesis, we present a solution to a subproblem within the learning by demonstration field, namely human-robot grasp mapping. Robot grasping of objects in a home or office environment is challenging problem. Programming by demonstration systems, can give important skills for aiding the robot in the grasping task. The thesis presents two techniques for human-robot grasp mapping, direct robot imitation from human demonstrator and intelligent grasp imitation. In intelligent grasp mapping, the robot takes the size and shape of the object into consideration, while for direct mapping, only the pose of the human hand is available. These are evaluated in a simulated environment on several robot platforms. The results show that knowing the object shape and size for a grasping task improves the robot precision and performance
Format: application/pdf
application/zip
Document access: http://hdl.handle.net/10256/7581
Language: eng
Collection: Enginyeria Tècnica. Informàtica de Sistemes (ETIS)
Rights: Attribution-NonCommercial-NoDerivs 3.0 Spain
Rights URI: http://creativecommons.org/licenses/by-nc-nd/3.0/es/
Subject: Intel·ligència artificial
Robòtica
Artificial intelligence
Robotics
Title: Intelligent task-level grasp mapping for robot control
Type: info:eu-repo/semantics/bachelorThesis
Repository: DUGiDocs

Subjects

Authors