¹Robert I. Citron,^2,3Peter Jenniskens,⁴Christopher Watkins,⁵Sravanthi Sinha,⁶Amar Shah,⁷Chedy Raissi,⁸Hadrien Devillepoix,²Jim Albers
Meteoritics & Planetary Science (in Press) Link to Article [https://doi.org/10.1111/maps.13663]
¹Department of Earth and Planetary Sciences, University of California, Davis, Davis, California, 95616 USA
²SETI Institute, Mountain View, California, 94043 USA
³NASA Ames Research Center, Moffett Field, California, 94035 USA
⁴Scientific Computing, Commonwealth Scientific and Industrial Research Organisation, Clayton, Victoria, 3181 Australia
⁵Holberton School of Software Engineering, San Francisco, California, 94111 USA
⁶Department of Engineering, Computational and Biological Learning, Cambridge University, Cambridge, CB2 1PZ UK
⁷Institut National de Recherche en Informatique et en Automatique, Villers-lès-Nancy, 54506 France
⁸Space Science & Technology Centre, School of Earth and Planetary Sciences, Curtin University, GPO Box U1987, Perth, Western Australia, 6845 Australia
Published by arrangement with John Wiley & Sons

The recovery of freshly fallen meteorites from tracked and triangulated meteors is critical to determining their source asteroid families. Even though our ability to locate meteorite falls continues to improve, the recovery of meteorites remains a challenge due to large search areas with terrain and vegetation obscuration. To improve the efficiency of meteorite recovery, we have tested the hypothesis that meteorites can be located using machine learning techniques and an autonomous drone. To locate meteorites autonomously, a quadcopter drone first conducts a grid survey acquiring top-down images of the strewn field from a low altitude. The drone-acquired images are then analyzed using a machine learning classifier to identify meteorite candidates for follow-up examination. Here, we describe a proof-of-concept meteorite classifier that deploys off-line a combination of different convolution neural networks to recognize meteorites from images taken by drones in the field. The system was implemented in a conceptual drone setup and tested in the suspected strewn field of a recent meteorite fall near Walker Lake, Nevada.