A cooperative behavior learning control of multi-robot using trace information

Tomofumi Ohshita, Ji Sun Shin, Michio Miyazaki, HeeHyol Lee

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

The distributed autonomous robotic system has superiority of robustness and adaptability to dynamical environment, however, the system requires the cooperative behavior mutually for optimality of the system. The acquisition of action by reinforcement learning is known as one of the approaches when the multi-robot works with cooperation mutually for a complex task. This paper deals with the transporting problem of the multi-robot using Q-learning algorithm in the reinforcement learning. When a robot carries luggage, we regard it as that the robot leaves a trace to the own migrational path, which trace has feature of volatility, and then, the other robot can use the trace information to help the robot, which carries luggage. To solve these problems on multi-agent reinforcement learning, the learning control method using stress antibody allotment reward is used. Moreover, we propose the trace information of the robot to urge cooperative behavior of the multi-robot to carry luggage to a destination in this paper. The effectiveness of the proposed method is shown by simulation.

Original languageEnglish
Title of host publicationProceedings of the 13th International Symposium on Artificial Life and Robotics, AROB 13th'08
Pages397-400
Number of pages4
Publication statusPublished - 2008
Event13th International Symposium on Artificial Life and Robotics, AROB 13th'08 - Oita
Duration: 2008 Jan 312008 Feb 2

Other

Other13th International Symposium on Artificial Life and Robotics, AROB 13th'08
CityOita
Period08/1/3108/2/2

Fingerprint

Robots
Reinforcement learning
Robustness (control systems)
Antibodies
Learning algorithms
Robotics

Keywords

  • Cooperative behavior
  • Multi-agent systems
  • Reinforcement learning
  • Stress antibody allotment reward

ASJC Scopus subject areas

  • Artificial Intelligence
  • Computer Vision and Pattern Recognition
  • Human-Computer Interaction

Cite this

Ohshita, T., Shin, J. S., Miyazaki, M., & Lee, H. (2008). A cooperative behavior learning control of multi-robot using trace information. In Proceedings of the 13th International Symposium on Artificial Life and Robotics, AROB 13th'08 (pp. 397-400)

A cooperative behavior learning control of multi-robot using trace information. / Ohshita, Tomofumi; Shin, Ji Sun; Miyazaki, Michio; Lee, HeeHyol.

Proceedings of the 13th International Symposium on Artificial Life and Robotics, AROB 13th'08. 2008. p. 397-400.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Ohshita, T, Shin, JS, Miyazaki, M & Lee, H 2008, A cooperative behavior learning control of multi-robot using trace information. in Proceedings of the 13th International Symposium on Artificial Life and Robotics, AROB 13th'08. pp. 397-400, 13th International Symposium on Artificial Life and Robotics, AROB 13th'08, Oita, 08/1/31.
Ohshita T, Shin JS, Miyazaki M, Lee H. A cooperative behavior learning control of multi-robot using trace information. In Proceedings of the 13th International Symposium on Artificial Life and Robotics, AROB 13th'08. 2008. p. 397-400
Ohshita, Tomofumi ; Shin, Ji Sun ; Miyazaki, Michio ; Lee, HeeHyol. / A cooperative behavior learning control of multi-robot using trace information. Proceedings of the 13th International Symposium on Artificial Life and Robotics, AROB 13th'08. 2008. pp. 397-400
@inproceedings{e5c989f4931f49e09386bf3933fb1aee,
title = "A cooperative behavior learning control of multi-robot using trace information",
abstract = "The distributed autonomous robotic system has superiority of robustness and adaptability to dynamical environment, however, the system requires the cooperative behavior mutually for optimality of the system. The acquisition of action by reinforcement learning is known as one of the approaches when the multi-robot works with cooperation mutually for a complex task. This paper deals with the transporting problem of the multi-robot using Q-learning algorithm in the reinforcement learning. When a robot carries luggage, we regard it as that the robot leaves a trace to the own migrational path, which trace has feature of volatility, and then, the other robot can use the trace information to help the robot, which carries luggage. To solve these problems on multi-agent reinforcement learning, the learning control method using stress antibody allotment reward is used. Moreover, we propose the trace information of the robot to urge cooperative behavior of the multi-robot to carry luggage to a destination in this paper. The effectiveness of the proposed method is shown by simulation.",
keywords = "Cooperative behavior, Multi-agent systems, Reinforcement learning, Stress antibody allotment reward",
author = "Tomofumi Ohshita and Shin, {Ji Sun} and Michio Miyazaki and HeeHyol Lee",
year = "2008",
language = "English",
isbn = "9784990288020",
pages = "397--400",
booktitle = "Proceedings of the 13th International Symposium on Artificial Life and Robotics, AROB 13th'08",

}

TY - GEN

T1 - A cooperative behavior learning control of multi-robot using trace information

AU - Ohshita, Tomofumi

AU - Shin, Ji Sun

AU - Miyazaki, Michio

AU - Lee, HeeHyol

PY - 2008

Y1 - 2008

N2 - The distributed autonomous robotic system has superiority of robustness and adaptability to dynamical environment, however, the system requires the cooperative behavior mutually for optimality of the system. The acquisition of action by reinforcement learning is known as one of the approaches when the multi-robot works with cooperation mutually for a complex task. This paper deals with the transporting problem of the multi-robot using Q-learning algorithm in the reinforcement learning. When a robot carries luggage, we regard it as that the robot leaves a trace to the own migrational path, which trace has feature of volatility, and then, the other robot can use the trace information to help the robot, which carries luggage. To solve these problems on multi-agent reinforcement learning, the learning control method using stress antibody allotment reward is used. Moreover, we propose the trace information of the robot to urge cooperative behavior of the multi-robot to carry luggage to a destination in this paper. The effectiveness of the proposed method is shown by simulation.

AB - The distributed autonomous robotic system has superiority of robustness and adaptability to dynamical environment, however, the system requires the cooperative behavior mutually for optimality of the system. The acquisition of action by reinforcement learning is known as one of the approaches when the multi-robot works with cooperation mutually for a complex task. This paper deals with the transporting problem of the multi-robot using Q-learning algorithm in the reinforcement learning. When a robot carries luggage, we regard it as that the robot leaves a trace to the own migrational path, which trace has feature of volatility, and then, the other robot can use the trace information to help the robot, which carries luggage. To solve these problems on multi-agent reinforcement learning, the learning control method using stress antibody allotment reward is used. Moreover, we propose the trace information of the robot to urge cooperative behavior of the multi-robot to carry luggage to a destination in this paper. The effectiveness of the proposed method is shown by simulation.

KW - Cooperative behavior

KW - Multi-agent systems

KW - Reinforcement learning

KW - Stress antibody allotment reward

UR - http://www.scopus.com/inward/record.url?scp=78449264179&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=78449264179&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:78449264179

SN - 9784990288020

SP - 397

EP - 400

BT - Proceedings of the 13th International Symposium on Artificial Life and Robotics, AROB 13th'08

ER -