UPDATE (12/2020): We published a Matlab script to easily reproduce the results of the DREGON paper (IROS 2018 [1]) using the latest version of the MBSS source localization toolbox (V2.0). Check out the Code and Algorithms page.

UPDATE (06/2019): Now that the IEEE SPCup 2019 is over, all the DREGON data are back online, as well as all the data from the competition and from the bonus task.

Welcome to the DREGON dataset webpage! DREGON, for DRone EGonoise and localizatiON, is a publically available dataset of annotated sounds recorded with an 8-channel microphone array embedded into a quadrotor UAV (Unmanned Aerial Vehicle). It aims at promoting research in UAV-embedded sound source localization, notably for the application of semi-autonomous search-and-rescue with drones. The data are free to use for academic and educational purpose.

Check out the following presentation video associated to our IROS 2018 paper [1] :

[1] DREGON: Dataset and Methods for UAV-Embedded Sound Source Localization, Martin Strauss, Pol Mordel, Victor Miguet and Antoine Deleforge, IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) 2018.