Peter Stone's Selected Publications

Classified by TopicClassified by Publication TypeSorted by DateSorted by First Author Last NameClassified by Funding Source


Learning to Order Objects Using Haptic and Proprioceptive Exploratory Behaviors

Jivko Sinapov, Priyanka Khante, Maxwell Svetlik, and Peter Stone. Learning to Order Objects Using Haptic and Proprioceptive Exploratory Behaviors. In Proceedings of the 25th International Joint Conference on Artificial Intelligence (IJCAI), July 2016.

Download

[PDF]6.6MB  [slides.pdf]5.2MB  

Abstract

This paper proposes a novel framework that enablesa robot to learn ordinal object relations. While mostrelated work focuses on classifying objects into dis-crete categories, such approaches cannot learn ob-ject properties (e.g., weight, height, size, etc.) thatare context-specific and relative to other objects.To address this problem, we propose that a robotshould learn to order objects based on ordinal ob-ject relations. In our experiments, the robot ex-plored a set of 32 objects that can be ordered bythree properties: height, weight, and width. Next,the robot used unsupervised learning to discovermultiple ways that the objects can be ordered basedon the haptic and proprioceptive perceptions de-tected while exploring the objects. Following, therobot’s model was presented with labeled object se-ries, allowing it to ground the three ordinal relationsin terms of how similar they are to the orders dis-covered during the unsupervised stage. Finally, thegrounded models were used to recognize whethernew object series were ordered by any of the threeproperties as well as to correctly insert additionalobjects into an existing series.

BibTeX Entry

@InProceedings{IJCAI16-sinapov,
  author = {Jivko Sinapov and Priyanka Khante and Maxwell Svetlik and Peter Stone},
  title = {Learning to Order Objects Using Haptic and Proprioceptive Exploratory Behaviors},
  booktitle = {Proceedings of the 25th International Joint Conference on Artificial
  Intelligence (IJCAI)},
  location = {New York City, USA},
  month = {July},
  year = {2016},
  abstract = {This paper proposes a novel framework that enables
a robot to learn ordinal object relations. While most
related work focuses on classifying objects into dis-
crete categories, such approaches cannot learn ob-
ject properties (e.g., weight, height, size, etc.)  that
are  context-specific  and  relative  to  other  objects.
To address this problem,  we propose that a robot
should learn to order objects based on ordinal ob-
ject  relations.   In  our  experiments,  the  robot  ex-
plored a set of 32 objects that can be ordered by
three properties:  height, weight, and width.  Next,
the  robot  used  unsupervised  learning  to  discover
multiple ways that the objects can be ordered based
on  the  haptic  and  proprioceptive  perceptions  de-
tected while exploring the objects.  Following, the
robot’s model was presented with labeled object se-
ries, allowing it to ground the three ordinal relations
in terms of how similar they are to the orders dis-
covered during the unsupervised stage. Finally, the
grounded models were used to recognize whether
new object series were ordered by any of the three
properties as well as to correctly insert additional
objects into an existing series.
  },
}

Generated by bib2html.pl (written by Patrick Riley ) on Mon Aug 19, 2019 13:01:02