Paper Abstract and Keywords |
Presentation |
2013-03-14 09:30
Improvement of Limited Action Selection in Reinforcement Learning Based Search Method for Ships' Courses Tomohiro Tanigawa, Takeshi Kamio (Hiroshima City Univ.), Kunihiko Mitsubori (Takushoku Univ.), Takahiro Tanaka (Japan Coast Guard Acad.), Hisato Fujisaka, Kazuhisa Haeiwa (Hiroshima City Univ.) NLP2012-144 |
Abstract |
(in Japanese) |
(See Japanese page) |
(in English) |
Ship transport is important. Multi ship course problem has been treared in the engineering related to ships. But the optimality of courses and the interaction between maneuvering actions have not been sufficiently discussed yet. So we have been designed multi-agent reinforcement learning system (MARLS). However, we have been confirmed by computer simulations that the resulting route is to avoid unnecessary. In this paper, we propose a limited action selection that correct to avoid unnecessary. |
Keyword |
(in Japanese) |
(See Japanese page) |
(in English) |
multi-agent reinforcement learning system / multi-ship course problem / prior knowledge / navigation rules / / / / |
Reference Info. |
IEICE Tech. Rep., vol. 112, no. 487, NLP2012-144, pp. 1-5, March 2013. |
Paper # |
NLP2012-144 |
Date of Issue |
2013-03-07 (NLP) |
ISSN |
Print edition: ISSN 0913-5685 Online edition: ISSN 2432-6380 |
Copyright and reproduction |
All rights are reserved and no part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopy, recording, or any information storage and retrieval system, without permission in writing from the publisher. Notwithstanding, instructors are permitted to photocopy isolated articles for noncommercial classroom use without fee. (License No.: 10GA0019/12GB0052/13GB0056/17GB0034/18GB0034) |
Download PDF |
NLP2012-144 |