Multi-robot systems with agent-based reinforcement learning: evolution, opportunities and challenges

Erfu Yang, Dongbing Gu

Research output: Contribution to journalArticle

7 Citations (Scopus)

Abstract

Multi-agent reinforcement learning for multi-robot systems is a challenging issue in both robotics and artificial intelligence. With the ever increasing interests in theoretical researches and practical applications, currently there have been a lot of efforts towards providing good solutions to this challenge. However, there are still many difficulties in scaling up multi-agent reinforcement learning to multi-robot systems. This paper presents a survey on the evolution, opportunities and challenges of applying agent-based reinforcement learning to multi-robot systems. After reviewing some important advances in this field, some challenging problems and promising research directions are focused on. A concluding remark is made from the perspectives of the authors.

LanguageEnglish
Pages271-286
Number of pages16
JournalInternational Journal of Modelling, Identification and Control
Volume6
Issue number4
DOIs
Publication statusPublished - 12 May 2009

Fingerprint

Multi-robot Systems
Reinforcement learning
Reinforcement Learning
Multiagent Learning
Robots
Artificial intelligence
Robotics
Artificial Intelligence
Scaling

Keywords

  • approximation and generalisation
  • fuzzy logic
  • MRSs
  • multi-agent systems
  • multi-robot systems
  • reinforcement learning
  • stochastic games
  • survey

Cite this

@article{d13270d346d34d33bc2b568651471fa4,
title = "Multi-robot systems with agent-based reinforcement learning: evolution, opportunities and challenges",
abstract = "Multi-agent reinforcement learning for multi-robot systems is a challenging issue in both robotics and artificial intelligence. With the ever increasing interests in theoretical researches and practical applications, currently there have been a lot of efforts towards providing good solutions to this challenge. However, there are still many difficulties in scaling up multi-agent reinforcement learning to multi-robot systems. This paper presents a survey on the evolution, opportunities and challenges of applying agent-based reinforcement learning to multi-robot systems. After reviewing some important advances in this field, some challenging problems and promising research directions are focused on. A concluding remark is made from the perspectives of the authors.",
keywords = "approximation and generalisation, fuzzy logic, MRSs, multi-agent systems, multi-robot systems, reinforcement learning, stochastic games, survey",
author = "Erfu Yang and Dongbing Gu",
year = "2009",
month = "5",
day = "12",
doi = "10.1504/IJMIC.2009.024735",
language = "English",
volume = "6",
pages = "271--286",
journal = "International Journal of Modelling, Identification and Control",
issn = "1746-6172",
number = "4",

}

TY - JOUR

T1 - Multi-robot systems with agent-based reinforcement learning

T2 - International Journal of Modelling, Identification and Control

AU - Yang, Erfu

AU - Gu, Dongbing

PY - 2009/5/12

Y1 - 2009/5/12

N2 - Multi-agent reinforcement learning for multi-robot systems is a challenging issue in both robotics and artificial intelligence. With the ever increasing interests in theoretical researches and practical applications, currently there have been a lot of efforts towards providing good solutions to this challenge. However, there are still many difficulties in scaling up multi-agent reinforcement learning to multi-robot systems. This paper presents a survey on the evolution, opportunities and challenges of applying agent-based reinforcement learning to multi-robot systems. After reviewing some important advances in this field, some challenging problems and promising research directions are focused on. A concluding remark is made from the perspectives of the authors.

AB - Multi-agent reinforcement learning for multi-robot systems is a challenging issue in both robotics and artificial intelligence. With the ever increasing interests in theoretical researches and practical applications, currently there have been a lot of efforts towards providing good solutions to this challenge. However, there are still many difficulties in scaling up multi-agent reinforcement learning to multi-robot systems. This paper presents a survey on the evolution, opportunities and challenges of applying agent-based reinforcement learning to multi-robot systems. After reviewing some important advances in this field, some challenging problems and promising research directions are focused on. A concluding remark is made from the perspectives of the authors.

KW - approximation and generalisation

KW - fuzzy logic

KW - MRSs

KW - multi-agent systems

KW - multi-robot systems

KW - reinforcement learning

KW - stochastic games

KW - survey

UR - http://www.scopus.com/inward/record.url?scp=65149100003&partnerID=8YFLogxK

U2 - 10.1504/IJMIC.2009.024735

DO - 10.1504/IJMIC.2009.024735

M3 - Article

VL - 6

SP - 271

EP - 286

JO - International Journal of Modelling, Identification and Control

JF - International Journal of Modelling, Identification and Control

SN - 1746-6172

IS - 4

ER -