University of Cincinnati team uses sound to flag drone damage for NASA

The project began in 2024 out of NASA's interest in detecting motor problems before a drone takes off, according to Chawla. That origin explains why the current work stays on the ground: the team is focusing on pre-flight tests NASA already runs, rather than on in-flight monitoring. The recordings of chipped or cracked propellers and failing motors are converted into data for ground station computers, so the check is meant to catch mechanical faults through changes in sound rather than through visual inspection alone.

The faults the team describes are not only visible damage. Chawla notes that grit and other debris can get into motor bearings and cause a motor to seize, which makes an acoustic check a complement to looking the aircraft over. Beyond the preflight setting, the students envision an autonomous drone that relies on sound alone to detect damage to its propellers and other mechanical systems during flight, returning to base before a catastrophic failure.

Busse ties the idea to fully autonomous drone-in-a-box systems, where a weatherproof docking station handles launch, landing, charging, data upload and mission scheduling without an on-site operator. In that setup, sensors in the box could provide preflight checks, he said. The project is set to be presented at the Modelling, Estimation and Control Conference on October 25 in Phoenix.