Pedestrians who wear headphones or earbuds to listen to music, text, or talk on smartphones while walking, putting themselves at risk by tuning out the traffic around them.
The Pedestrian Audio Wearable System (PAWS) is a low-cost, headset-based, wearable platform that combines five MEMS microphones, signal processing, and feature extraction electronics as well as machine learning classifiers running on a smartphone to help detect and locate imminent dangers such as approaching cars and warn pedestrians in real time using audio/visual feedback on the user’s smartphone. A microcontroller-based front-end hardware platform consists of commercial off-the-shelf (COTS) components embedded into a standard headset that collects four channels of audio from the microphones positioned on the headset. Four of the MEMS microphones at the user’s left and right ear, back of the head, and chest provide relevant information about the sound source’s location. The front-end hardware synchronously acquires analog signals from these microphones and locally extracts acoustic features that are used by a smartphone application. The front-end hardware is battery-powered and uses its own set of microphones for sound processing. It does not interact with the speakers or microphone of the headset. As such, a user does not experience any degradation in sound or microphone quality of the headset. The mechanism will be designed so that people will recognize the alert and respond quickly.
Data Science Institute, Columbia University, New York, NY
Known as “twalking,” texting and walking is not without its dangers. Headphone-wearing pedestrians often can’t hear the auditory cues — horns, shouts, or the sound of approaching cars — that signal imminent harm. As a result, the number of injuries and deaths caused by twalking in the U.S. has tripled in the past seven years. Last year, pedestrian deaths in the U.S. were at their highest level since 1990.
The team is now testing its design both in the lab and on the streets of New York City, which is known for its congestion and its cacophony of sounds. Once refined, the technology will be commercialized and mass-produced by a commercial company.