• English
    • norsk
  • English 
    • English
    • norsk
  • Login
View Item 
  •   All institutions
  • Norges teknisk-naturvitenskapelige universitet
  • Publikasjoner fra CRIStin - NTNU
  • View Item
  •   All institutions
  • Norges teknisk-naturvitenskapelige universitet
  • Publikasjoner fra CRIStin - NTNU
  • View Item
JavaScript is disabled for your browser. Some features of this site may not work without it.

AUV Pipeline Following using Reinforcement Learning

Fjerdingen, Sigurd Aksnes; Kyrkjebø, Erik; Transeth, Aksel Andreas
Chapter
Accepted version
Thumbnail
View/Open
SINTEF+S16087.pdf (161.4Kb)
Permanent link
http://hdl.handle.net/11250/2489643
Issue date
2010
Share
Metadata
Show full item record
Collections
  • Publikasjoner fra CRIStin - NTNU [7701]
  • Institutt for teknisk kybernetikk [1332]
Abstract
This paper analyzes the application of several reinforcement learning techniques for continuous state and action spaces to pipeline following for an autonomous underwater vehicle (AUV). Continuous space SARSA is compared to the actor-critic CACLA algorithm, and is also extended into a supervised reinforcement learning architecture. A novel exploration method using the skew-normal stochastic distribution is proposed, and evidence towards advantages in the case of tabula rasa exploration is presented. Results are validated on a realistic simulator of the AUV, and confirm the applicability of reinforcement learning to optimize pipeline following behavior.
Publisher
VDE Verlag GmbH

Contact Us

Search NORA
Powered by DSpace software

Service from BIBSYS
 

 

Browse this CollectionIssue DateAuthorsTitlesSubjectsDocument TypesJournalsBrowse all ArchivesArchives & CollectionsIssue DateAuthorsTitlesSubjectsDocument TypesJournals

My Account

Login

Statistics

Google Analytics StatisticsView Usage Statistics

Contact Us

Search NORA
Powered by DSpace software

Service from BIBSYS