Forget Car Crash Detection: Pixel Phones Can Also Help Prevent Subway Accidents

Eric Zeman / Android Authority

Your smartphone is utterly packed to the brim with sensor of all kinds. Some of those are out in the open, like the cameras our phones use to image their surroundings, while others are hidden within, like the accelerometers that tell our phones what direction we’re holding them. Collectively, these sensors can be used by clever software to deliver all sorts of useful features, like how Pixel phones are able to detect when we’re in a car crash and summon emergency assistance. But now we’re hearing about how those same sensors are keeping some much larger vehicles safe, with New York City’s Metropolitan Transportation Authority (MTA) tapping Pixel phones for some help with subway maintenance.

The MTA’s got a big job, responsible for keeping the population of the country’s biggest city moving around without clogging the streets with an impossible number of cars. But running a subway system this expansive means there’s a lot of track that you’ve got to keep safe and operational — some 665 miles of it.

Normally, looking for track issues in need of repair either requires manually walking those miles and miles of tracks, or scanning them with “train geometry cars,” packed with sensors able to identify problem areas. But recently, Wired reports, the MTA partnered up with Google to see if Pixel phones could maybe replace some of those expensive scanner cars.

Rather than using dedicated equipment to only scan subway lines a few times a year, more affordable hardware could allow the MTA to deploy it more broadly, gathering data with higher frequency and potentially spotting problems even earlier. And as smartphones are basically commodity devices now, strapping a Pixel phone to a subway car makes for a cheap and easy way to augment it with sensors.

The TrackInspect system combines vibration monitoring and audio recordings with location data. Once that’s run through some AI analysis, those bump and bangs can be use to identify areas where tracks may be in need of service. During the four-month test period, the Pixel phones correctly caught 92% of the problems later confirmed by human inspectors.

With so much riding on subway safety, no one’s rushing to hand over full maintenance responsibility to Pixel phones and AI algorithms just yet, and rules currently require that trained inspectors oversee this kind of work. But should the MTA choose to adopt something like TrackInspect long-term, it could be a powerful tool for making those inspectors’ jobs a whole lot easier — and keep the subway safe for all of us.

Got a tip? Talk to us! Email our staff at [email protected]. You can stay anonymous or get credit for the info, it’s your choice.