Summary

Proceedings of the 2013 International Symposium on Nonlinear Theory and its Applications

2013

Session Number:C4L-C

Session:

Number:487

Modified Multi-Agent Reinforcement Learning System to Find Ships' Courses

Tomohiro Tanigawa,  Takeshi Kamio,  Kunihiko Mitsubori,  Takahiro Tanaka,  Hisato Fujisaka,  Kazuhisa Haeiwa,  

pp.487-490

Publication Date:

Online ISSN:2188-5079

DOI:10.15248/proc.2.487

PDF download (563.4KB)

Summary:
Although the ship transportation is important for low cost mass transit, the optimality of ships' courses and the interaction between maneuvering actions have not been sufficiently discussed yet. In order to brisk up these discussions, we have developed multi-agent reinforcement learning system (MARLS) to find ships' courses. However, our previous MARLS cannot search the courses, considering both efficiency and safety.
In this paper, we modify our MARLS to overcome this problem. From numerical experiments, we have confirmed that our modified MARLS can get more efficient courses than our previous MARLS, although both of them have similar performance with regard to safety.

References:

[1] K. Mitsubori, T. Kamio, and T. Tanaka, “Finding the shortest course of a ship based on reinforcement learning algorithm,” Journal of Japan Institute of Navigation, 110, pp.9-18, 2004.

[2] K. Mitsubori, T. Kamio, and T. Tanaka, “Finding the course and collision avoidance based on reinforcement learning algorithm,” NAVIGATION, 170, pp.26-31, 2009 (in Japanese).

[3] T. Kamio, S. Sugeo, K. Mitsubori, T. Tanaka, C. J. Ahn, H. Fujisaka, and K. Haeiwa, “A reinforcement learning approach to course decision of ships under navigation rules,” Proc. of NOLTA, pp.141-144, 2009.

[4] T. Kamio, K. Mitsubori, T. Tanaka, H. Fujisaka, K. Haeiwa, “Effects of prior knowledge on multi-agent reinforcement leaning system to find courses of ships,” Australian Journal of Intelligent Information Processing Systems, vol.12, no.2, pp.18-23, 2010.

[5] International Maritime Organization, “International regulations for preventing collisions at sea,” 1972.

[6] M. Inaishi, H. Kondo, A. Kawaguchi “Multiple Ship Clusters Behavior Simulation” Journal of Japan Institute of Navigation, 110, pp.1-7, 2004.

[7] M. Inaishi, H. Kondo, M. Kondo, A. Kawaguchi “Marine Traffic Simulation Using Ship Agent Clusters : Northbound Traffic in Tokyo Bay” Journal of Japan Institute of Navigation, 115, pp.11-16, 2006.

[8] T. I. Fossen, “Guidance and control of ocean vehicle,” John Wiley & Sons Ltd., pp.172-174, 1994.

[9] M. Fuchi, S. Usui, S. Fujimoto, K. Hirono and T. Mochida, “Differences in allowable distance between ships due to ship's specific type,” Japan Ergonomics Society, vol.46, pp.144-145, 2010 (in Japanese).