Show simple item record

dc.contributor.authorPolvara, R
dc.contributor.authorPatacchiola, M
dc.contributor.authorsharma, sanjay
dc.contributor.authorWan, Jian
dc.contributor.authorManning, Andrew
dc.contributor.authorSutton, R
dc.contributor.authorCangelosi, Angelo
dc.contributor.authorIEEE,
dc.date.accessioned2019-03-06T10:16:58Z
dc.date.available2019-03-06T10:16:58Z
dc.date.issued2018-09-03
dc.identifier.isbn9781538613535
dc.identifier.issn2373-6720
dc.identifier.urihttp://hdl.handle.net/10026.1/13441
dc.description.abstract

The autonomous landing of an unmanned aerial vehicle (UAV) is still an open problem. Previous work focused on the use of hand-crafted geometric features and sensor-data fusion for identifying a fiducial marker and guide the UAV toward it. In this article we propose a method based on deep reinforcement learning that only requires low-resolution images coming from a down looking camera in order to drive the vehicle. The proposed approach is based on a hierarchy of Deep Q-Networks (DQNs) that are used as high-end control policy for the navigation in different phases. We implemented various technical solutions, such as the combination of vanilla and double DQNs trained using a form of prioritized buffer replay that separates experiences in multiple containers. The optimal control policy is learned without any human supervision, providing the agent with a sparse reward feedback indicating the success or failure of the landing. The results show that the quadrotor can autonomously land on a large variety of simulated environments and with relevant noise, proving that the underline DQNs are able to generalise effectively on unseen scenarios. Furthermore, it was proved that in some conditions the network outperformed human pilots.

dc.format.extent115-123
dc.language.isoen
dc.publisherIEEE
dc.titleToward End-to-End Control for UAV Autonomous Landing via Deep Reinforcement Learning
dc.typeconference
dc.typeConference Proceeding
plymouth.author-urlhttp://gateway.webofknowledge.com/gateway/Gateway.cgi?GWVersion=2&SrcApp=PARTNER_APP&SrcAuth=LinksAMR&KeyUT=WOS:000454860700016&DestLinkType=FullRecord&DestApp=ALL_WOS&UsrCustomerID=11bb513d99f797142bcfeffcc58ea008
plymouth.date-start2018-06-12
plymouth.date-finish2018-06-15
plymouth.volume00
plymouth.conference-name2018 International Conference on Unmanned Aircraft Systems (ICUAS)
plymouth.publication-statusPublished
plymouth.journal2018 International Conference on Unmanned Aircraft Systems (ICUAS)
dc.identifier.doi10.1109/icuas.2018.8453449
plymouth.organisational-group/Plymouth
plymouth.organisational-group/Plymouth/Faculty of Science and Engineering
plymouth.organisational-group/Plymouth/Faculty of Science and Engineering/School of Biological and Marine Sciences
plymouth.organisational-group/Plymouth/Faculty of Science and Engineering/School of Engineering, Computing and Mathematics
plymouth.organisational-group/Plymouth/REF 2021 Researchers by UoA
plymouth.organisational-group/Plymouth/REF 2021 Researchers by UoA/UoA07 Earth Systems and Environmental Sciences
plymouth.organisational-group/Plymouth/REF 2021 Researchers by UoA/UoA12 Engineering
plymouth.organisational-group/Plymouth/Research Groups
plymouth.organisational-group/Plymouth/Research Groups/Institute of Health and Community
plymouth.organisational-group/Plymouth/Research Groups/Marine Institute
plymouth.organisational-group/Plymouth/Users by role
plymouth.organisational-group/Plymouth/Users by role/Academics
plymouth.organisational-group/Plymouth/Users by role/Researchers in ResearchFish submission
dcterms.dateAccepted2018-04-14
dc.rights.embargodate9999-12-31
dc.rights.embargoperiodNot known
rioxxterms.versionofrecord10.1109/icuas.2018.8453449
rioxxterms.licenseref.urihttp://www.rioxx.net/licenses/all-rights-reserved
rioxxterms.licenseref.startdate2018-09-03
rioxxterms.typeConference Paper/Proceeding/Abstract


Files in this item

Thumbnail
Thumbnail

This item appears in the following Collection(s)

Show simple item record


All items in PEARL are protected by copyright law.
Author manuscripts deposited to comply with open access mandates are made available in accordance with publisher policies. Please cite only the published version using the details provided on the item record or document. In the absence of an open licence (e.g. Creative Commons), permissions for further reuse of content should be sought from the publisher or author.
Theme by 
Atmire NV