Presentation 2017-10-06
Influence of Updating Reference Courses on Reinforcement Learning to Search Ships' Courses
Takeshi Kamio, Takahiro Tanaka, Kunihiko Mitsubori, Hisato Fujisaka,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Deciding efficient and safe courses of ships before actual navigation is very important. We have developed multi-agent reinforcement learning system (MARLS) to search ships’ courses as a useful tool to discuss the appropriateness of courses and the interaction between maneuvering actions. Recently, we have considered the introduction of reference courses to MARLS to increase the value of MARLS as the tool. In this research, we propose a more simple method to trace the reference courses than Ref. [12] and evaluate the influence of updating reference courses on MARLS through the numerical simulations.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Multi-ship course problem / Reinforcement learning / Multi-agent system / Limited action selection / Reference courses
Paper # CAS2017-38,NLP2017-63
Date of Issue 2017-09-28 (CAS, NLP)

Conference Information
Committee NLP / CAS
Conference Date 2017/10/5(2days)
Place (in Japanese) (See Japanese page)
Place (in English) Machinaka Campus Nagaoka
Topics (in Japanese) (See Japanese page)
Topics (in English) etc.
Chair Masaharu Adachi(Tokyo Denki Univ.) / Mitsuru Hiraki(Renesas)
Vice Chair Norikazu Takahashi(Okayama Univ.) / Hideaki Okazaki(Shonan Inst. of Tech.)
Secretary Norikazu Takahashi(Nagaoka Univ. of Tech.) / Hideaki Okazaki(Hiroshima Inst. of Tech.)
Assistant Toshihiro Tachibana(Shonan Inst. of Tech.) / Masayuki Kimura(Kyoto Univ.) / Yohei Nakamura(Hitachi)

Paper Information
Registration To Technical Committee on Nonlinear Problems / Technical Committee on Circuits and Systems
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Influence of Updating Reference Courses on Reinforcement Learning to Search Ships' Courses
Sub Title (in English)
Keyword(1) Multi-ship course problem
Keyword(2) Reinforcement learning
Keyword(3) Multi-agent system
Keyword(4) Limited action selection
Keyword(5) Reference courses
1st Author's Name Takeshi Kamio
1st Author's Affiliation Hiroshima City University(Hiroshima City Univ.)
2nd Author's Name Takahiro Tanaka
2nd Author's Affiliation Japan Coast Guard Academy(Japan Coast Guard Academy)
3rd Author's Name Kunihiko Mitsubori
3rd Author's Affiliation Takushoku University(Takushoku Univ.)
4th Author's Name Hisato Fujisaka
4th Author's Affiliation Hiroshima City University(Hiroshima City Univ.)
Date 2017-10-06
Paper # CAS2017-38,NLP2017-63
Volume (vol) vol.117
Number (no) CAS-225,NLP-226
Page pp.pp.75-80(CAS), pp.75-80(NLP),
#Pages 6
Date of Issue 2017-09-28 (CAS, NLP)