An Espressif ESP32 Sensor Package Turns a Small COTS Drone Into a Volcano-Monitoring Little-RAVEN

Researchers at the Johannes Gutenberg-Universität Mainz, working with colleagues at Heidelberg University and the National Institute of Geophysics and Volcanology (INGV) in Catania, have turned compact backpack-portable drones into sensor systems for monitoring volcanoes — and potentially predicting when an eruption might happen.

"Small, easily transportable drones are an essential prerequisite if we want to get to isolated or difficult-to-access volcanic sites and suitably track the activity there," explains lead author Niklas Karbach of the project, which saw commercial off-the-shelf (COTS) DJI Mavic 3 drones equipped with a custom sensor payload deployed on the island of Volcano, at the southern boundary of the Aeolian Islands.

The team aren't the first to try drones as a means of monitoring volcanic activity, but prior efforts have been hampered by the size of the drones used to carry often-bulky sensor packages — and the corresponding cost and complexity in transporting the drones to the volcanoes. The DJI Mavic 3, by contrast, weighs less than two pounds — even when equipped with its sensor payload, turning it into what the researchers called the little Remote-Controlled Aircraft for Volcanic Emission Analysis, or "little-RAVEN."

The sensor payload itself is based on an Espressif ESP32 microcontroller running at 240MHz and with 4MB of flash storage, connected to an Alpasense SO2-B4 sulphur dioxide (SO₂) sensor, an ELT Sensors S300 carbon dioxide (CO₂) sensor, a Bosch Sensortec BME280 temperature, pressure, and humidity sensor, a GPS module, and a micro pump connected to a B+B Sensors FLW-122 flow sensor. The sensor payload includes its own 1.3Ah battery, separate to the drone's battery pack.

"We need to obtain real-time data on sulfur dioxide levels as this lets us know when we are actually in contact with the volcanic plume," Thorsten Hoffmann, professor and project lead, explains, "something that readily moves over time in response to atmospheric factors. The localization of a plume by visual means alone from a distance of several kilometers is practically impossible."

Testing of the little-RAVEN platform proved it capable of finding and measuring volcanic plumes, even in the hands of a lightly-trained pilot — and of providing data that can be used to predict when an eruption may occur, with far higher measurement frequencies than existing systems. "In addition," the researchers conclude, "the easy-to-operate and safe systems allow monitoring tasks at comparatively low costs, which is an argument not to be neglected in volcano monitoring for geological/governmental institutions in economically weak regions."

The team's work has been published under open-access terms in the journal Scientific Reports.