Show simple item record

dc.contributor.authorMüller, Matthias
dc.contributor.authorCasser, Vincent
dc.contributor.authorSmith, Neil
dc.contributor.authorMichels, Dominik L.
dc.contributor.authorGhanem, Bernard
dc.date.accessioned2019-03-28T07:05:27Z
dc.date.available2017-12-28T07:32:16Z
dc.date.available2018-11-28T14:01:50Z
dc.date.available2019-03-28T07:05:27Z
dc.date.issued2019-01-29
dc.identifier.citationMüller M, Casser V, Smith N, Michels DL, Ghanem B (2019) Teaching UAVs to Race: End-to-End Regression of Agile Controls in Simulation. Physics of Solid Surfaces: 11–29. Available: http://dx.doi.org/10.1007/978-3-030-11012-3_2.
dc.identifier.issn1615-1925
dc.identifier.doi10.1007/978-3-030-11012-3_2
dc.identifier.urihttp://hdl.handle.net/10754/626557
dc.description.abstractAutomating the navigation of unmanned aerial vehicles (UAVs) in diverse scenarios has gained much attention in recent years. However, teaching UAVs to fly in challenging environments remains an unsolved problem, mainly due to the lack of training data. In this paper, we train a deep neural network to predict UAV controls from raw image data for the task of autonomous UAV racing in a photo-realistic simulation. Training is done through imitation learning with data augmentation to allow for the correction of navigation mistakes. Extensive experiments demonstrate that our trained network (when sufficient data augmentation is used) outperforms state-of-the-art methods and flies more consistently than many human pilots. Additionally, we show that our optimized network architecture can run in real-time on embedded hardware, allowing for efficient on-board processing critical for real-world deployment. From a broader perspective, our results underline the importance of extensive data augmentation techniques to improve robustness in end-to-end learning setups.
dc.description.sponsorshipThis work was supported by the King Abdullah University of Science and Technology (KAUST) Office of Sponsored Research through the Visual Computing Center (VCC) funding.
dc.publisherSpringer Nature
dc.relation.urlhttps://link.springer.com/chapter/10.1007%2F978-3-030-11012-3_2
dc.rightsArchived with thanks to Physics of Solid Surfaces
dc.titleTeaching UAVs to Race: End-to-End Regression of Agile Controls in Simulation
dc.typeConference Paper
dc.contributor.departmentVisual Computing Center (VCC)
dc.contributor.departmentComputer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division
dc.contributor.departmentElectrical Engineering Program
dc.identifier.journalPhysics of Solid Surfaces
dc.conference.date2018-09-08 to 2018-09-14
dc.conference.name15th European Conference on Computer Vision, ECCV 2018
dc.conference.locationMunich, DEU
dc.eprint.versionPost-print
dc.identifier.arxivid1708.05884
kaust.personMüller, Matthias
kaust.personCasser, Vincent
kaust.personSmith, Neil
kaust.personMichels, Dominik L.
kaust.personGhanem, Bernard
refterms.dateFOA2018-06-14T09:22:37Z
dc.date.published-online2019-01-29
dc.date.published-print2019
dc.date.posted2017-08-19


Files in this item

Thumbnail
Name:
Teaching_UAVs_to_Race__ECCVW_18_.pdf
Size:
5.269Mb
Format:
PDF
Description:
Accepted Manuscript

This item appears in the following Collection(s)

Show simple item record

VersionItemEditorDateSummary

*Selected version