Researchers Are Now Teaching Drones to Detect Screaming People

Unmanned aircraft have been long implemented into search and rescue operations. And while drones can easily accomplish tasks of surveilling vast regions and accessing hard-to-reach locations, there's still one issue that limits the machine's performance: finding missing persons or injured people in disaster-struck areas where there's no visibility.
The image shows the SJI Air S2 drone 1 photo
Photo: DJI
In a new study, a team of researchers from German detailed how an array of microphones installed on a drone combined with advanced processing techniques could save someone's life. The project was presented Tuesday, June 8th, at the Acoustical Society of America's annual conference.

During the conference, Macarena Varela, a researcher from Fraunhofer FKIE, explains how survivors usually make noise so the rescuers can find them faster. Therefore, the team tried to design an accurate acoustic system that could be mounted on a drone and used to locate potential victims.

Of course, the system had to be able to differentiate between ambient noise, aircraft noise, and human screaming. To filter out the unnecessary sounds, various impulsive sounds, drone noise, and speech noise were recorded and saved in a database.

Then the system was tested using what is called the Crow's Nest Array (CNA), a microphone array that includes a navigation system, a modular sensor system, wireless communication, and uses a frequency range from 5 Hz to 100 Hz.

Together with modern signal processing techniques, the team successfully achieved the detection of precise locations of certain sounds made by missing people.

Because drones cannot carry heavy payloads, the team had to minimize the components in quantity, weight, and size. Therefore, the microphone array was made up of tiny digital Micro-Electro-Mechanical-Systems (MEMS) microphones.

To date, the team has conducted different experiments, including open field ones which proved to be successful. Researchers plan to integrate an additional condenser microphone with a wider frequency range to provide a more precise signal for detection and categorization.
If you liked the article, please follow us:  Google News icon Google News Youtube Instagram
About the author: Florina Spînu
Florina Spînu profile photo

Florina taught herself how to drive in a Daewoo Tico (a rebadged Suzuki Alto kei car) but her first "real car" was a VW Golf. When she’s not writing about cars, drones or aircraft, Florina likes to read anything related to space exploration and take pictures in the middle of nature.
Full profile


Would you like AUTOEVOLUTION to send you notifications?

You will only receive our top stories