Thumbnail
Access Restriction
Subscribed

Author Tamosiunaite, Minija ♦ Asfour, Tamim ♦ Wörgötter, Florentin
Source SpringerLink
Content type Text
Publisher Springer-Verlag
File Format PDF
Copyright Year ©2009
Language English
Subject Domain (in DDC) Technology ♦ Medicine & health
Subject Keyword Reinforcement learning ♦ Function approximation ♦ Robot control ♦ Neurosciences ♦ Computer Application in Life Sciences ♦ Neurobiology ♦ Bioinformatics ♦ Statistical Physics, Dynamical Systems and Complexity
Abstract Reinforcement learning methods can be used in robotics applications especially for specific target-oriented problems, for example the reward-based recalibration of goal directed actions. To this end still relatively large and continuous state-action spaces need to be efficiently handled. The goal of this paper is, thus, to develop a novel, rather simple method which uses reinforcement learning with function approximation in conjunction with different reward-strategies for solving such problems. For the testing of our method, we use a four degree-of-freedom reaching problem in 3D-space simulated by a two-joint robot arm system with two DOF each. Function approximation is based on 4D, overlapping kernels (receptive fields) and the state-action space contains about 10,000 of these. Different types of reward structures are being compared, for example, reward-on- touching-only against reward-on-approach. Furthermore, forbidden joint configurations are punished. A continuous action space is used. In spite of a rather large number of states and the continuous action space these reward/punishment strategies allow the system to find a good solution usually within about 20 trials. The efficiency of our method demonstrated in this test scenario suggests that it might be possible to use it on a real robot for problems where mixed rewards can be defined in situations where other types of learning might be difficult.
ISSN 03401200
Age Range 18 to 22 years ♦ above 22 year
Educational Use Research
Education Level UG and PG
Learning Resource Type Article
Publisher Date 2009-02-20
Publisher Place Berlin/Heidelberg
e-ISSN 14320770
Journal Biological Cybernetics
Volume Number 100
Issue Number 3
Page Count 12
Starting Page 249
Ending Page 260


Open content in new tab

   Open content in new tab
Source: SpringerLink