LiU Electronic Press
Download:
File size:
486 kb
Format:
application/pdf
Author:
Kleiner, Alexander (University of Freiburg)
Dietl, M. (University of Freiburg) (Foundations of Artificial Intelligence)
Nebel, Bernhard (University of Freiburg) (Foundations of Artificial Intelligence)
Title:
Towards a Life-Long Learning Soccer Agent
Publication type:
Conference paper (Refereed)
Language:
English
In:
In RoboCup 2002: Robot Soccer World Cup VI
Volume:
2752
Pages:
126-134
Year of publ.:
2002
URI:
urn:nbn:se:liu:diva-72554
Permanent link:
http://urn.kb.se/resolve?urn=urn:nbn:se:liu:diva-72554
Subject category:
Robotics
Project:
Artificial Intelligence & Integrated Computer Systems
Abstract(en) :

One problem in robotic soccer (and in robotics in general) is to adapt skills and the overall behavior to a changing environment and to hardware improvements. We applied hierarchical reinforcement learning in an SMDP framework learning on all levels simultaneously. As our experiments show, learning simultaneously on the skill level and on the skill selection level is advantageous since it allows for a smooth adaption to a changing environment. Furthermore, the skills we trained turn also out to be quite competitive when run on the real robotic players of the players of our CS Freiburg team.

Available from:
2011-11-28
Created:
2011-11-28
Last updated:
2011-12-01
Statistics:
0 hits
FILE INFORMATION
File size:
486 kb
Mimetype:
application/pdf
Type:
fulltext
Statistics:
0 hits
Version:
Authorʼs version