A Learning Approach to Multi-robot Task Allocation with Priority Constraints and Uncertainty

Multi-robot task allocation has an important impact on the efficiency of multi-robot collaboration. For single-shot allocation without complicated constraints, some exact algorithms and heuristic algorithms can find the optimal solution efficiently. However, considering the priority constraints and...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:2022 IEEE International Conference on Industrial Technology (ICIT) s. 1 - 8
Hlavní autoři: Deng, Fuqin, Huang, Huanzhao, Fu, Lanhui, Yue, Hongwei, Zhang, Jianmin, Wu, Zexiao, Lam, Tin Lun
Médium: Konferenční příspěvek
Jazyk:angličtina
Vydáno: IEEE 22.08.2022
Témata:
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Abstract Multi-robot task allocation has an important impact on the efficiency of multi-robot collaboration. For single-shot allocation without complicated constraints, some exact algorithms and heuristic algorithms can find the optimal solution efficiently. However, considering the priority constraints and uncertain execution time of robots for multiple times of allocation in an approximate dynamic programming environment, traditional methods such as heuristic algorithms have limited performance. To obtain better performance, we propose a method based on deep reinforcement learning. Specifically, we first use the directed acyclic graph to describe the priority relationship between tasks. Then we propose a graph neural network with a hierarchical attention mechanism to extract the characteristics of the task groups. Finally, we design the policy network to solve the approximate dynamic programming problem of multi-robot task allocation. Through training on the dataset of a given environment, the policy network can gradually refine the decision-making process by reinforcement learning. Experiment results show that the proposed modeling and solving method can find better solutions than existing heuristic algorithms. Furthermore, the learned strategy can be directly applied in other untrained environments with superior performance.
AbstractList Multi-robot task allocation has an important impact on the efficiency of multi-robot collaboration. For single-shot allocation without complicated constraints, some exact algorithms and heuristic algorithms can find the optimal solution efficiently. However, considering the priority constraints and uncertain execution time of robots for multiple times of allocation in an approximate dynamic programming environment, traditional methods such as heuristic algorithms have limited performance. To obtain better performance, we propose a method based on deep reinforcement learning. Specifically, we first use the directed acyclic graph to describe the priority relationship between tasks. Then we propose a graph neural network with a hierarchical attention mechanism to extract the characteristics of the task groups. Finally, we design the policy network to solve the approximate dynamic programming problem of multi-robot task allocation. Through training on the dataset of a given environment, the policy network can gradually refine the decision-making process by reinforcement learning. Experiment results show that the proposed modeling and solving method can find better solutions than existing heuristic algorithms. Furthermore, the learned strategy can be directly applied in other untrained environments with superior performance.
Author Deng, Fuqin
Fu, Lanhui
Zhang, Jianmin
Lam, Tin Lun
Yue, Hongwei
Huang, Huanzhao
Wu, Zexiao
Author_xml – sequence: 1
  givenname: Fuqin
  surname: Deng
  fullname: Deng, Fuqin
  organization: Wuyi University,School of Intelligent Manufacturing,Jiangmen,Guangdong,China,529020
– sequence: 2
  givenname: Huanzhao
  surname: Huang
  fullname: Huang, Huanzhao
  organization: Wuyi University,School of Intelligent Manufacturing,Jiangmen,Guangdong,China,529020
– sequence: 3
  givenname: Lanhui
  surname: Fu
  fullname: Fu, Lanhui
  organization: Wuyi University,School of Intelligent Manufacturing,Jiangmen,Guangdong,China,529020
– sequence: 4
  givenname: Hongwei
  surname: Yue
  fullname: Yue, Hongwei
  organization: Wuyi University,School of Intelligent Manufacturing,Jiangmen,Guangdong,China,529020
– sequence: 5
  givenname: Jianmin
  surname: Zhang
  fullname: Zhang, Jianmin
  email: zjm99_2001@126.com
  organization: Wuyi University,School of Intelligent Manufacturing,Jiangmen,Guangdong,China,529020
– sequence: 6
  givenname: Zexiao
  surname: Wu
  fullname: Wu, Zexiao
  organization: The 3irobotix Co.,Ltd,Shenzhen,Guangdong,China,518000
– sequence: 7
  givenname: Tin Lun
  surname: Lam
  fullname: Lam, Tin Lun
  email: tllam@cuhk.edu.cn
  organization: The Shenzhen Institute of Artificial Intelligence and Robotics for Society,Shenzhen,Guangdong,China,518000
BookMark eNo1j8tKxDAYRiPoQsd5A8G8QGsu7UyyLMVLoaKLzk4Y_tycYE1KGpG-vSOjq8O3OXznCp2HGCxCt5SUlBJ517XdUIkN4SUjjJWUEMIEqc_QWm4F3TJBqayEvERvDe4tpODDO26mKUXQB5wjfv4asy9SVDHjAeYP3Ixj1JB9DPjb5wN-TT4mnxfcxjDnBD7kGUMweBe0Tfl3L9fowsE42_UfV2j3cD-0T0X_8ti1TV_4441caMEBDBOMOWq1IbUTTDlnQVGpK9hoDZJrfQzjUANQ6TShTihrDFGV4XyFbk5eb63dT8l_Qlr2_9H8B6oTVBM
ContentType Conference Proceeding
DBID 6IE
6IL
CBEJK
RIE
RIL
DOI 10.1109/ICIT48603.2022.10002805
DatabaseName IEEE Electronic Library (IEL) Conference Proceedings
IEEE Xplore POP ALL
IEEE Xplore All Conference Proceedings
IEEE Electronic Library (IEL)
IEEE Proceedings Order Plans (POP All) 1998-Present
DatabaseTitleList
Database_xml – sequence: 1
  dbid: RIE
  name: IEEE Electronic Library (IEL)
  url: https://ieeexplore.ieee.org/
  sourceTypes: Publisher
DeliveryMethod fulltext_linktorsrc
EISBN 9781728119489
1728119480
EndPage 8
ExternalDocumentID 10002805
Genre orig-research
GrantInformation_xml – fundername: Shenzhen Peacock Plan
  funderid: 10.13039/501100012234
GroupedDBID 6IE
6IL
CBEJK
RIE
RIL
ID FETCH-LOGICAL-i119t-c83aad2822f1ecd05f82bffeab19c4a6cca93cc1103a5aa19fc01f8bedd0b4d33
IEDL.DBID RIE
IngestDate Thu Jan 18 11:13:56 EST 2024
IsPeerReviewed false
IsScholarly false
Language English
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-i119t-c83aad2822f1ecd05f82bffeab19c4a6cca93cc1103a5aa19fc01f8bedd0b4d33
PageCount 8
ParticipantIDs ieee_primary_10002805
PublicationCentury 2000
PublicationDate 2022-Aug.-22
PublicationDateYYYYMMDD 2022-08-22
PublicationDate_xml – month: 08
  year: 2022
  text: 2022-Aug.-22
  day: 22
PublicationDecade 2020
PublicationTitle 2022 IEEE International Conference on Industrial Technology (ICIT)
PublicationTitleAbbrev ICIT
PublicationYear 2022
Publisher IEEE
Publisher_xml – name: IEEE
Score 1.8119875
Snippet Multi-robot task allocation has an important impact on the efficiency of multi-robot collaboration. For single-shot allocation without complicated constraints,...
SourceID ieee
SourceType Publisher
StartPage 1
SubjectTerms Approximation algorithms
Deep reinforcement learning
Dynamic programming
Graph neural network
Graph neural networks
Heuristic algorithms
Multi-robot task allocation
Reinforcement learning
Training
Uncertainty
Title A Learning Approach to Multi-robot Task Allocation with Priority Constraints and Uncertainty
URI https://ieeexplore.ieee.org/document/10002805
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1NSwMxEA1aPHhSseI3OXhNu9lss8mxFIsFKT200INQJl9SlF1pU6H_3iTdKh48eAshJDBDeJPJvHkIPXAnBBMZI6XKDSm0AgIcDOFaMkmFda5IROHncjwW87mcNGT1xIWx1qbiM9uJw_SXb2q9iamyLk3JoNix9LAs-Y6s1dRs0Ux2R4PRNGoqsfDsy_POfvUv3ZQEG8OTfx54ito_BDw8-YaWM3Rgq3P00sdNN9RX3G9agWNf48ShJata1R5PYf2G--8RoaLFcUyzhp2WddSow1GdM2lC-DWGyuBZOCdVBPhtG82Gj9PBE2nEEciSUumJFgzAxCJQR602Wc-JXDlnQVGpC-DBM5JpHQzDoAdApdMZdUJZYzJVGMYuUKuqK3uJsBUmgDhAuM5QqIwLbUpOtZUh1stDeHOF2tE0i49d_4vF3irXf8zfoOPogJh5zfNb1PKrjb1DR_rTL9er--S1L5kenHM
linkProvider IEEE
linkToHtml http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1NSwMxEA1SBT2pWPHbHLym3STbbXIsxdJiLT1soQehTL6kKLvSbgX_vUm6Kh48eAshJDBDeJPJvHkI3WVOCC4STrqKGZJqBQQyMCTTkksqrHNpJAqPu5OJmM_ltCarRy6MtTYWn9lWGMa_fFPqTUiVtWlMBoWOpbudNGXJlq5VV23RRLZH_VEeVJW4f_gx1vpa_0s5JQLH4PCfRx6h5g8FD0-_weUY7djiBD31cN0P9Rn36mbguCpxZNGSVanKCuewfsG914BRweY4JFr9TssyqNThoM8ZVSGqNYbC4Jk_J9YEVB9NNBvc5_0hqeURyJJSWREtOIAJZaCOWm2SjhNMOWdBUalTyLxvJNfaG4ZDB4BKpxPqhLLGJCo1nJ-iRlEW9gxhK4yHcQB_oSFVSSa06WZUW-mjPeYDnHPUDKZZvG07YCy-rHLxx_wt2h_mj-PFeDR5uEQHwRkhD8vYFWpUq429Rnv6vVquVzfRg5_pPJ-6
openUrl ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Abook&rft.genre=proceeding&rft.title=2022+IEEE+International+Conference+on+Industrial+Technology+%28ICIT%29&rft.atitle=A+Learning+Approach+to+Multi-robot+Task+Allocation+with+Priority+Constraints+and+Uncertainty&rft.au=Deng%2C+Fuqin&rft.au=Huang%2C+Huanzhao&rft.au=Fu%2C+Lanhui&rft.au=Yue%2C+Hongwei&rft.date=2022-08-22&rft.pub=IEEE&rft.spage=1&rft.epage=8&rft_id=info:doi/10.1109%2FICIT48603.2022.10002805&rft.externalDocID=10002805