The UT Austin Team

Layered Learning on a Physical Robot

Peggy Fidelman and Peter Stone. Layered Learning on a Physical Robot. In Under Review. , February 2005.
Some videos of the robot referenced in the paper.

Download

[PDF]151.8kB  [postscript]2.5MB  

Abstract

Layered learning is a general hierarchical machine learning paradigm that leverages a given task decomposition to learn complex tasks efficiently. Though it has been validated previously in simulation, this paper presents the first application of layered learning on a physical robot. In particular, it enables the learning of a high-level grasping behavior that relies on a gait which itself must be learned. All learning is done autonomously onboard a commercially available Sony Aibo robot, with no human intervention other than battery changes. We demonstrate that our approach makes it possible to quickly learn both a fast gait and a reliable grasping behavior which, in combination, significantly outperform our best hand-tuned solution.

BibTeX Entry

@InProceedings(IROS2005-chinpinch,
author="Peggy Fidelman and Peter Stone",
title="Layered Learning on a Physical Robot",
booktitle="Under Review. ",
month="February",year="2005",
abstract={
Layered learning is a general hierarchical machine
learning paradigm that leverages a given task
decomposition to learn complex tasks efficiently.
Though it has been validated previously in
simulation, this paper presents the first
application of layered learning on a physical robot.
In particular, it enables the learning of a
high-level grasping behavior that relies on a gait
which itself must be learned. All learning is done
autonomously onboard a commercially available Sony
Aibo robot, with no human intervention other than
battery changes. We demonstrate that our approach
makes it possible to quickly learn both a fast gait
and a reliable grasping behavior which, in
combination, significantly outperform our best
hand-tuned solution.
},
)

Valid CSS!
Valid XHTML 1.0!