Policy Gradient Reinforcement Learning for Fast Quadrupedal Locomotion
Policy Gradient Reinforcement Learning for Fast Quadrupedal Locomotion.
Nate Kohl
and Peter Stone.
In Proceedings of the IEEE International Conference
on Robotics and Automation, May 2004.
Some videos
of walking robots referenced in the paper.
ICRA 2004
Download
[PDF]301.5kB [postscript]3.2MB
Abstract
This paper presents a machine learning approach to optimizing a quadrupedal trot gait for forward speed. Given a parameterized walk designed for a specific robot, we propose using a form of policy gradient reinforcement learning to automatically search the set of possible parameters with the goal of finding the fastest possible walk. We implement and test our approach on a commercially available quadrupedal robot platform, namely the Sony Aibo robot. After about three hours of learning, all on the physical robots and with no human intervention other than to change the batteries, the robots achieved the fastest walk known for the Aibo, significantly outperforming a variety of existing hand-coded and learned solutions.
BibTeX Entry
@InProceedings(icra04,
author="Nate Kohl and Peter Stone",
title="Policy Gradient Reinforcement Learning for Fast Quadrupedal Locomotion",
year="2004",month="May",
booktitle="Proceedings of the {IEEE} International Conference on Robotics and Automation",
abstract={
This paper presents a machine learning approach to
optimizing a quadrupedal trot gait for forward speed.
Given a parameterized walk designed for a specific
robot, we propose using a form of policy gradient
reinforcement learning to automatically search the
set of possible parameters with the goal of finding
the fastest possible walk. We implement and test our
approach on a commercially available quadrupedal
robot platform, namely the Sony Aibo robot. After
about three hours of learning, all on the physical
robots and with no human intervention other than to
change the batteries, the robots achieved the fastest
walk known for the Aibo, significantly outperforming
a variety of existing hand-coded and learned
solutions.
},
)
