File Download
Supplementary

Conference Paper: Flying through a narrow gap using neural network: an end-to-end planning and control approach

TitleFlying through a narrow gap using neural network: an end-to-end planning and control approach
Authors
KeywordsDeep Learning in Robotics and Automation
Motion and Path Planning
Issue Date2019
PublisherIEEE/RSJ.
Citation
IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS): Robots Connecting People, Macau, China, 4-8 November 2019 How to Cite?
AbstractIn this paper, we investigate the problem of enabling a drone to fly through a tilted narrow gap, without a traditional planning and control pipeline. To this end, we propose an end-to-end policy network, which imitates from the traditional pipeline and is fine-tuned using reinforcement learning. Unlike previous works which plan dynamical feasible trajectories using motion primitives and track the generated trajectory by a geometric controller, our proposed method is an end-to-end approach which takes the flight scenario as input and directly outputs thrust-attitude control commands for the quadrotor. Key contributions of our paper are: 1) presenting an imitatereinforce training framework. 2) flying through a narrow gap using an end-to-end policy network, showing that learning based method can also address the highly dynamic control problem as the traditional pipeline does (see attached video1). 3) propose a robust imitation of an optimal trajectory generator using multilayer perceptrons. 4) show how reinforcement learning can improve the performance of imitation learning, and the potential to achieve higher performance over the model-based method.
DescriptionWeBT3 Regular session, L1-R3: Learning for Motion and Path Planning II - Paper WeBT3.2
Persistent Identifierhttp://hdl.handle.net/10722/274125

 

DC FieldValueLanguage
dc.contributor.authorLin, J-
dc.contributor.authorWang, L-
dc.contributor.authorGao, F-
dc.contributor.authorShen, S-
dc.contributor.authorZhang, F-
dc.date.accessioned2019-08-18T14:55:35Z-
dc.date.available2019-08-18T14:55:35Z-
dc.date.issued2019-
dc.identifier.citationIEEE/RSJ International Conference on Intelligent Robots and Systems (IROS): Robots Connecting People, Macau, China, 4-8 November 2019-
dc.identifier.urihttp://hdl.handle.net/10722/274125-
dc.descriptionWeBT3 Regular session, L1-R3: Learning for Motion and Path Planning II - Paper WeBT3.2-
dc.description.abstractIn this paper, we investigate the problem of enabling a drone to fly through a tilted narrow gap, without a traditional planning and control pipeline. To this end, we propose an end-to-end policy network, which imitates from the traditional pipeline and is fine-tuned using reinforcement learning. Unlike previous works which plan dynamical feasible trajectories using motion primitives and track the generated trajectory by a geometric controller, our proposed method is an end-to-end approach which takes the flight scenario as input and directly outputs thrust-attitude control commands for the quadrotor. Key contributions of our paper are: 1) presenting an imitatereinforce training framework. 2) flying through a narrow gap using an end-to-end policy network, showing that learning based method can also address the highly dynamic control problem as the traditional pipeline does (see attached video1). 3) propose a robust imitation of an optimal trajectory generator using multilayer perceptrons. 4) show how reinforcement learning can improve the performance of imitation learning, and the potential to achieve higher performance over the model-based method.-
dc.languageeng-
dc.publisherIEEE/RSJ.-
dc.relation.ispartofIEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)-
dc.subjectDeep Learning in Robotics and Automation-
dc.subjectMotion and Path Planning-
dc.titleFlying through a narrow gap using neural network: an end-to-end planning and control approach-
dc.typeConference_Paper-
dc.identifier.emailZhang, F: fuzhang@hku.hk-
dc.identifier.authorityZhang, F=rp02460-
dc.description.naturepostprint-
dc.identifier.hkuros301104-

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats