Collective behavior acquisition of real robotic swarms using deep reinforcement learning

Toshiyuki Yasuda; Kazuhiro Ohkura

doi:10.1109/IRC.2018.00038

Collective behavior acquisition of real robotic swarms using deep reinforcement learning

Toshiyuki Yasuda, Kazuhiro Ohkura

Mechanical Engineering

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

12 Scopus citations

Abstract

Swarm robotic systems are a type of multi-robot systems, in which robots operate without any form of centralized control. The most popular approach for SRS is the so-called ad hoc or behavior-based approach; desired collective behavior is obtained by manually by designing the behavior of individual robot in advance. On the other hand, in the principled or automatic design approach, a certain general methodology for developing appropriate collective behavior is adopted. This paper investigates a deep reinforcement learning approach to collective behavior acquisition of swarm robotics systems. Robots are expected to collect information in parallel and share their experience for accelerating the learning. We conduct real swarm robot experiments and evaluate the learning performance in a scenario where robots consecutively travel between two landmarks.

Original language	English
Title of host publication	Proceedings - 2nd IEEE International Conference on Robotic Computing, IRC 2018
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	179-180
Number of pages	2
ISBN (Electronic)	9781538646519
DOIs	https://doi.org/10.1109/IRC.2018.00038
State	Published - 2018/04/02
Event	2nd IEEE International Conference on Robotic Computing, IRC 2018 - Laguna Hills, United States Duration: 2018/01/31 → 2018/02/02

Publication series

Name	Proceedings - 2nd IEEE International Conference on Robotic Computing, IRC 2018
Volume	2018-January

Conference

Conference	2nd IEEE International Conference on Robotic Computing, IRC 2018
Country/Territory	United States
City	Laguna Hills
Period	2018/01/31 → 2018/02/02

Keywords

Deep reinforcement learning
Experience sharing
Swarm robotics

ASJC Scopus subject areas

Control and Optimization
Artificial Intelligence
Software

Access to Document

10.1109/IRC.2018.00038

Cite this

Yasuda, T., & Ohkura, K. (2018). Collective behavior acquisition of real robotic swarms using deep reinforcement learning. In Proceedings - 2nd IEEE International Conference on Robotic Computing, IRC 2018 (pp. 179-180). (Proceedings - 2nd IEEE International Conference on Robotic Computing, IRC 2018; Vol. 2018-January). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/IRC.2018.00038

@inproceedings{4bff6e3e4c4e473fb9d30cf46cd3a5f0,

title = "Collective behavior acquisition of real robotic swarms using deep reinforcement learning",

abstract = "Swarm robotic systems are a type of multi-robot systems, in which robots operate without any form of centralized control. The most popular approach for SRS is the so-called ad hoc or behavior-based approach; desired collective behavior is obtained by manually by designing the behavior of individual robot in advance. On the other hand, in the principled or automatic design approach, a certain general methodology for developing appropriate collective behavior is adopted. This paper investigates a deep reinforcement learning approach to collective behavior acquisition of swarm robotics systems. Robots are expected to collect information in parallel and share their experience for accelerating the learning. We conduct real swarm robot experiments and evaluate the learning performance in a scenario where robots consecutively travel between two landmarks.",

keywords = "Deep reinforcement learning, Experience sharing, Swarm robotics",

author = "Toshiyuki Yasuda and Kazuhiro Ohkura",

note = "Publisher Copyright: {\textcopyright} 2018 IEEE.; 2nd IEEE International Conference on Robotic Computing, IRC 2018 ; Conference date: 31-01-2018 Through 02-02-2018",

year = "2018",

month = apr,

day = "2",

doi = "10.1109/IRC.2018.00038",

language = "英語",

series = "Proceedings - 2nd IEEE International Conference on Robotic Computing, IRC 2018",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "179--180",

booktitle = "Proceedings - 2nd IEEE International Conference on Robotic Computing, IRC 2018",

}

Yasuda, T & Ohkura, K 2018, Collective behavior acquisition of real robotic swarms using deep reinforcement learning. in Proceedings - 2nd IEEE International Conference on Robotic Computing, IRC 2018. Proceedings - 2nd IEEE International Conference on Robotic Computing, IRC 2018, vol. 2018-January, Institute of Electrical and Electronics Engineers Inc., pp. 179-180, 2nd IEEE International Conference on Robotic Computing, IRC 2018, Laguna Hills, United States, 2018/01/31. https://doi.org/10.1109/IRC.2018.00038

Collective behavior acquisition of real robotic swarms using deep reinforcement learning. / Yasuda, Toshiyuki; Ohkura, Kazuhiro.
Proceedings - 2nd IEEE International Conference on Robotic Computing, IRC 2018. Institute of Electrical and Electronics Engineers Inc., 2018. p. 179-180 (Proceedings - 2nd IEEE International Conference on Robotic Computing, IRC 2018; Vol. 2018-January).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Collective behavior acquisition of real robotic swarms using deep reinforcement learning

AU - Yasuda, Toshiyuki

AU - Ohkura, Kazuhiro

PY - 2018/4/2

Y1 - 2018/4/2

N2 - Swarm robotic systems are a type of multi-robot systems, in which robots operate without any form of centralized control. The most popular approach for SRS is the so-called ad hoc or behavior-based approach; desired collective behavior is obtained by manually by designing the behavior of individual robot in advance. On the other hand, in the principled or automatic design approach, a certain general methodology for developing appropriate collective behavior is adopted. This paper investigates a deep reinforcement learning approach to collective behavior acquisition of swarm robotics systems. Robots are expected to collect information in parallel and share their experience for accelerating the learning. We conduct real swarm robot experiments and evaluate the learning performance in a scenario where robots consecutively travel between two landmarks.

AB - Swarm robotic systems are a type of multi-robot systems, in which robots operate without any form of centralized control. The most popular approach for SRS is the so-called ad hoc or behavior-based approach; desired collective behavior is obtained by manually by designing the behavior of individual robot in advance. On the other hand, in the principled or automatic design approach, a certain general methodology for developing appropriate collective behavior is adopted. This paper investigates a deep reinforcement learning approach to collective behavior acquisition of swarm robotics systems. Robots are expected to collect information in parallel and share their experience for accelerating the learning. We conduct real swarm robot experiments and evaluate the learning performance in a scenario where robots consecutively travel between two landmarks.

KW - Deep reinforcement learning

KW - Experience sharing

KW - Swarm robotics

UR - http://www.scopus.com/inward/record.url?scp=85049612029&partnerID=8YFLogxK

U2 - 10.1109/IRC.2018.00038

DO - 10.1109/IRC.2018.00038

M3 - 会議への寄与

AN - SCOPUS:85049612029

T3 - Proceedings - 2nd IEEE International Conference on Robotic Computing, IRC 2018

SP - 179

EP - 180

BT - Proceedings - 2nd IEEE International Conference on Robotic Computing, IRC 2018

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2nd IEEE International Conference on Robotic Computing, IRC 2018

Y2 - 31 January 2018 through 2 February 2018

ER -

Yasuda T, Ohkura K. Collective behavior acquisition of real robotic swarms using deep reinforcement learning. In Proceedings - 2nd IEEE International Conference on Robotic Computing, IRC 2018. Institute of Electrical and Electronics Engineers Inc. 2018. p. 179-180. (Proceedings - 2nd IEEE International Conference on Robotic Computing, IRC 2018). doi: 10.1109/IRC.2018.00038

Collective behavior acquisition of real robotic swarms using deep reinforcement learning

Abstract

Publication series

Conference

Keywords

ASJC Scopus subject areas

Access to Document

Fingerprint

Cite this