Paper Info
Title
An Affordance Sensitive System for Robot to Human Object Handover.
Abstract
One of the most important characteristics that needs to be taken into account while designing advanced human-robot interaction systems is the ability of the robot to behave in a socially acceptable way that is comfortable for humans. This paper presents a novel system for robot to human object handover that maximizes user’s convenience while receiving the object. The object is delivered to the receiving partner such that the most appropriate part is oriented towards him/her. The system has been developed supporting all the necessary phases of the handover task, including object recognition, people detection, robot motion planning, and automatic detection of user’s grasp. Moreover, voice recognition and text-to-speech have been integrated to enable natural object selection in environments that include multiple objects. The experimental setup consists of a six degree of freedom robot arm equipped with a two-finger gripper and an eye-in-hand laser scanner for object recognition, as well as a fixed range sensor for people and grasp detection. A user study has been conducted to assess the usability of the system and verify whether novice users can successfully accomplish a handover task with the system. The user study has confirmed that the proposed solution allows a more comfortable handover than a system disregarding object orientation.
Year
DOI
Venue
2014
10.1007/s12369-014-0241-3
I. J. Social Robotics
Keywords
Field
DocType
Human-robot interaction,Object handover,Grasp affordances
Computer vision,Robotic arm,GRASP,Simulation,Usability,Psychology,Artificial intelligence,Robot,Affordance,Handover,Human–robot interaction,Cognitive neuroscience of visual object recognition
Journal
Volume
Issue
ISSN
6
4
1875-4791
Citations 
PageRank 
References 
10
0.74
19
Authors
3
Name
Order
Citations
PageRank
Jacopo Aleotti125929.76
Vincenzo Micelli2232.26
Stefano Caselli331436.32