By Topic

Understanding Object Weight from Human and Humanoid Lifting Actions

Sign In

Cookies must be enabled to login.After enabling cookies , please use refresh or reload or ctrl+f5 on the browser for the login options.

Formats Non-Member Member
$31 $13
Learn how you can qualify for the best price for this item!
Become an IEEE Member or Subscribe to
IEEE Xplore for exclusive pricing!
close button

puzzle piece

IEEE membership options for an individual and IEEE Xplore subscriptions for an organization offer the most affordable access to essential journal articles, conference papers, standards, eBooks, and eLearning courses.

Learn more about:

IEEE membership

IEEE Xplore subscriptions

4 Author(s)
Sciutti, A. ; Dept. of Robot., Brain & Cognitive Sci., Ist. Italiano di Tecnol., Genoa, Italy ; Patane, L. ; Nori, F. ; Sandini, G.

Humans are very good at interacting with each other. This natural ability depends, among other factors, on an implicit communication mediated by motion observation. By simple action observation we can easily infer not only the goal of an agent, but often also some “hidden” properties of the object he is manipulating, as its weight or its temperature. This implicit understanding is developed early in childhood and is supposedly based on a common motor repertoire between the cooperators. In this paper, we have investigated whether and under which conditions it is possible for a humanoid robot to foster the same kind of automatic communication, focusing on the ability to provide cues about object weight with action execution. We have evaluated on which action properties weight estimation is based in humans and we have accordingly designed a set of simple robotic lifting behaviors. Our results show that subjects can reach a performance in weight recognition from robot observation comparable to that obtained during human observation, with no need of training. These findings suggest that it is possible to design robot behaviors that are implicitly understandable by nonexpert partners and that this approach could be a viable path to obtain more natural human-robot collaborations.

Published in:

Autonomous Mental Development, IEEE Transactions on  (Volume:6 ,  Issue: 2 )