Show simple item record

dc.contributor.authorGondhalekar, Nahush Rameshen_US
dc.date.accessioned2017-08-04T08:00:43Z
dc.date.available2017-08-04T08:00:43Z
dc.date.issued2017-08-03
dc.identifier.othervt_gsexam:12330en_US
dc.identifier.urihttp://hdl.handle.net/10919/78667
dc.description.abstractWe study the problem of Reinforcement Learning (RL) for Unmanned Aerial Vehicle (UAV) navigation with the smallest number of real world samples possible. This work is motivated by applications of learning autonomous navigation for aerial robots in structural inspec- tion. A naive RL implementation suffers from curse of dimensionality in large continuous state spaces. Gaussian Processes (GPs) exploit the spatial correlation to approximate state- action transition dynamics or value function in large state spaces. By incorporating GPs in naive Q-learning we achieve better performance in smaller number of samples. The evalua- tion is performed using simulations with an aerial robot. We also present a Multi-Fidelity Reinforcement Learning (MFRL) algorithm that leverages Gaussian Processes to learn the optimal policy in a real world environment leveraging samples gathered from a lower fidelity simulator. In MFRL, an agent uses multiple simulators of the real environment to perform actions. With multiple levels of fidelity in a simulator chain, the number of samples used in successively higher simulators can be reduced.en_US
dc.format.mediumETDen_US
dc.publisherVirginia Techen_US
dc.rightsThis item is protected by copyright and/or related rights. Some uses of this item may be deemed fair and permitted by law even without permission from the rights holder(s), or the rights holder(s) may have licensed the work for use under certain conditions. For other uses you need to obtain permission from the rights holder(s).en_US
dc.subjectReinforcement Learningen_US
dc.subjectGaussian Processesen_US
dc.subjectUnmanned Aerial Vehicle Navigationen_US
dc.titleReinforcement Learning with Gaussian Processes for Unmanned Aerial Vehicle Navigationen_US
dc.typeThesisen_US
dc.contributor.departmentElectrical and Computer Engineeringen_US
dc.description.degreeMSen_US
thesis.degree.nameMSen_US
thesis.degree.levelmastersen_US
thesis.degree.grantorVirginia Polytechnic Institute and State Universityen_US
thesis.degree.disciplineComputer Engineeringen_US
dc.contributor.committeechairTokekar, Pratapen_US
dc.contributor.committeememberZeng, Haiboen_US
dc.contributor.committeememberAbbott, Amos Len_US


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record