Data

UAV Acoustic Localization Dataset: 24-Channel Beamformed Recordings of a DJI Air 3 Drone with Synchronized GPS Flight Logs

Western Sydney University
Rodriguez, Belman Jahir ; Chevtchenko, Sergio ; Afshar, Saeed
Viewed: [[ro.stat.viewed]] Cited: [[ro.stat.cited]] Accessed: [[ro.stat.accessed]]
ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Adc&rfr_id=info%3Asid%2FANDS&rft_id=info:doi10.26183/07nm-4p35&rft.title=UAV Acoustic Localization Dataset: 24-Channel Beamformed Recordings of a DJI Air 3 Drone with Synchronized GPS Flight Logs&rft.identifier=10.26183/07nm-4p35&rft.publisher=Western Sydney University&rft.description=This dataset provides real-world, open-field acoustic recordings of a DJI Air 3 drone captured with a custom ground-based 24-microphone array, synchronized with GPS flight telemetry. Recordings span multiple sessions across different days and two distinct outdoor locations, including drone-present and ambient no-drone segments used to calibrate noise-robust detection models. The dataset supports research in sound source localization (SSL) and Sound Event Localization and Detection (SELD), and was used to train and validate a U-Net-based model that reformulates DoA estimation as spherical semantic segmentation over delay-and-sum (DAS) beamformed acoustic energy maps. Dataset contents: Multichannel WAV audio recordings (24 channels, 48 kHz, four synchronized Zoom F6 recorders) of a DJI Air 3 drone in open-field flight, plus CSV flight-log files (GPS position, altitude, speed, heading at 100 ms resolution). Includes per-session alignment parameters (JSON) and reference-flight recordings used to calibrate the GPS-to-array coordinate frame, plus ambient no-drone background audio. Two of the four sessions also include a 360-degree reference video (Insta360 X4). Companion Python scripts are included so others can regenerate the labelled dataset used to train the model.&rft.creator=Rodriguez, Belman Jahir &rft.creator=Chevtchenko, Sergio &rft.creator=Afshar, Saeed &rft.date=2026&rft.coverage=Western Sydney, NSW&rft_rights=Copyright Western Sydney University&rft_rights=CC BY-SA 4.0: Attribution-Share Alike 4.0 International http://creativecommons.org/licenses/by-sa/4.0&rft_subject=sound source localization&rft_subject=direction of arrival estimation&rft_subject=microphone array&rft_subject=delay-and-sum beamforming&rft_subject=U-Net&rft_subject=SELD&rft_subject=acoustic imaging&rft_subject=semantic segmentation&rft_subject=drone acoustics&rft_subject=UAV detection&rft_subject=GPS-synchronized audio&rft_subject=open-field dataset&rft_subject=SDG 9 - Industry, Innovation and Infrastructure&rft_subject=SDG 16 - Peace, Justice and Strong Institutions&rft_subject=Signal processing&rft_subject=Communications engineering&rft_subject=ENGINEERING&rft_subject=Deep learning&rft_subject=Machine learning&rft_subject=INFORMATION AND COMPUTING SCIENCES&rft_subject=National security&rft_subject=Defence&rft_subject=DEFENCE&rft.type=dataset&rft.language=English Access the data
Cite Saved to MyRDA Save to MyRDA

Licence & Rights:

Open Licence view details
CC-BY-SA

CC BY-SA 4.0: Attribution-Share Alike 4.0 International
http://creativecommons.org/licenses/by-sa/4.0

Copyright Western Sydney University

Access:

Open view details

Open

Contact Information



Full description

This dataset provides real-world, open-field acoustic recordings of a DJI Air 3 drone captured with a custom ground-based 24-microphone array, synchronized with GPS flight telemetry. Recordings span multiple sessions across different days and two distinct outdoor locations, including drone-present and ambient "no-drone" segments used to calibrate noise-robust detection models. The dataset supports research in sound source localization (SSL) and Sound Event Localization and Detection (SELD), and was used to train and validate a U-Net-based model that reformulates DoA estimation as spherical semantic segmentation over delay-and-sum (DAS) beamformed acoustic energy maps.

Dataset contents: Multichannel WAV audio recordings (24 channels, 48 kHz, four synchronized Zoom F6 recorders) of a DJI Air 3 drone in open-field flight, plus CSV flight-log files (GPS position, altitude, speed, heading at 100 ms resolution). Includes per-session alignment parameters (JSON) and reference-flight recordings used to calibrate the GPS-to-array coordinate frame, plus ambient "no-drone" background audio. Two of the four sessions also include a 360-degree reference video (Insta360 X4). Companion Python scripts are included so others can regenerate the labelled dataset used to train the model.

Created: 2026-07-15

Data time period: 10 2024 to 31 03 2025

This dataset is part of a larger collection

Click to explore relationships graph
Help

Related People

Spatial Coverage And Location

text: Western Sydney, NSW

Subjects
Communications Engineering | Defence | Deep Learning | Defence | Engineering | GPS-synchronized audio | Information and Computing Sciences | Machine Learning | National Security | SDG 16 - Peace, Justice and Strong Institutions | SDG 9 - Industry, Innovation and Infrastructure | SELD | Signal Processing | U-Net | UAV detection | acoustic imaging | delay-and-sum beamforming | direction of arrival estimation | drone acoustics | microphone array | open-field dataset | semantic segmentation | sound source localization |

User Contributed Tags    

Login to tag this record with meaningful keywords to make it easier to discover

Identifiers
  • DOI : 10.26183/07NM-4P35
  • Local : research-data.westernsydney.edu.au/published/1f8cceb0801b11f189a547c836cf5bee
ACN 633 798 857