Towards monocular vision-based autonomous flight through deep reinforcement learning

Simple item page
dc.contributor.authorKim, Minwoo
dc.contributor.authorKim, Jongyun
dc.contributor.authorJung, Minjae
dc.contributor.authorOh, Hyondong
dc.date.accessioned2022-03-16T12:55:53Z
dc.date.available2022-03-16T12:55:53Z
dc.date.issued2022-03-09
dc.description.abstractThis paper proposes an obstacle avoidance strategy for small multi-rotor drones with a monocular camera using deep reinforcement learning. The proposed method is composed of two steps: depth estimation and navigation decision making. For the depth estimation step, a pre-trained depth estimation algorithm based on the convolutional neural network is used. On the navigation decision making step, a dueling double deep Q-network is employed with a well-designed reward function. The network is trained using the robot operating system and Gazebo simulation environment. To validate the performance and robustness of the proposed approach, simulations and real experiments have been carried out using a Parrot Bebop2 drone in various complex indoor environments. We demonstrate that the proposed algorithm successfully travels along the narrow corridors with the texture free walls, people, and boxes.en_UK
dc.identifier.citationKim M, Kim J, Jung M, Oh H. (2022) Towards monocular vision-based autonomous flight through deep reinforcement learning, Expert Systems with Applications, Volume 198, July 2022, Article number 116742en_UK
dc.identifier.issn0957-4174
dc.identifier.urihttps://doi.org/10.1016/j.eswa.2022.116742
dc.identifier.urihttps://dspace.lib.cranfield.ac.uk/handle/1826/17659
dc.language.isoenen_UK
dc.publisherElsevieren_UK
dc.rightsAttribution-NonCommercial-NoDerivatives 4.0 International*
dc.rights.urihttp://creativecommons.org/licenses/by-nc-nd/4.0/*
dc.subjectObstacle avoidanceen_UK
dc.subjectDepth estimationen_UK
dc.subjectVision-baseden_UK
dc.subjectDeep reinforcement learningen_UK
dc.subjectQ-learningen_UK
dc.subjectNavigation decision makingen_UK
dc.titleTowards monocular vision-based autonomous flight through deep reinforcement learningen_UK
dc.typeArticleen_UK

Files

Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
Monocular_vision-based_autonomous_flight-2022.pdf
Size:
23.83 MB
Format:
Adobe Portable Document Format
Description:
Download
License bundle
Now showing 1 - 1 of 1
No Thumbnail Available
Name:
license.txt
Size:
1.63 KB
Format:
Item-specific license agreed upon to submission
Description:
Download

Collections

Staff publications (SATM)