Summary
International Symposium on Nonlinear Theory and its Applications
2009
Session Number:A2L-D
Session:
Number:A2L-D4
A Reinforcement Learning Approach to Course Decision of Ships under Navigation Rules
Takeshi Kamio, Shohei Sugeo, Kunihiko Mitsubori, Takahiro Tanaka, Chang-Jun Ahn, Hisato Fujisaka, Kazuhisa Haeiwa,
pp.-
Publication Date:2009/10/18
Online ISSN:2188-5079
Summary:
The transportation by ship is very important in countries with seas and wide rivers. In the case of Japan, it accounts for about 40% of the domestic physical distribution and for 90% and more of the international physical distribution. Therefore, the course decision of ships is an important problem in the field of the marine engineering. However, the optimality of the course of ships and the interaction between the maneuvering actions of navigators have not been sufficiently discussed yet. We regard the multi agent reinforcement learning (RL), which is an important learning algorithm in the field of the artificial intelligence and the machine learning, as a useful tool to brisk up these discussions. In this paper, we propose the RL framework to decide the course of ships under the navigation rules.