A few Pixel smartphones have been routinely used on four New York City subway cars since September of last year, and it was for a purpose that would not just save the service countless funds but also avoid potentially life-threatening situations. According to the latest report, the Metropolitan Transportation Authority (MTA) and Google were experimenting with how these devices could prevent problems in subway cars thanks to their advanced internal sensors, negating the need to install the significantly more expensive equipment.
During the four-month testing period, Google’s Pixel smartphones leveraged experimental technology called TrackInspect
Traditionally, problems in subway cars or the infrastructure are inspected by humans, who have to walk 665 miles of the New York City transit while keeping their eyes open for anything unusual, such as broken rails or water damage. There are also ‘train geometry cars’ equipped with a bevy of sensors that capture and upload crucial data concerning the subway’s infrastructure. However, what if there was an inexpensive way of carrying out all of these tasks? The Google Public Sector might be working on a solution.
This division deployed experimental technology called TrackInspect, and when used by Pixel devices, can detect audio, vibration and location data. This information can then be used to train AI prediction models, which can go a long way in assisting humans during the inspection work. During the four-month period, Google’s handsets successfully identified 92 percent of the defect locations, which were later confirmed by the inspection team.
This approach can become a stepping stone for something far more advanced, with MTA’s President Demetrius Crichlow believing that a ‘modernized’ system can be created that identifies and initiates fixes for subways. As reported by Wired, Crichlow says that the goal is to identify the issues and eliminate them before they become a major problem and start disrupting the service. There are approximately 3.7 million daily commuters, meaning that these defects can make a difference for thousands of travelers who use the service for work or school.
TrackInspect reportedly combined 335 million sensor readings and 1,200 hours of audio, along with New York City Transit’s database of track defects, to train around 200 individual prediction models to find issues. The collaboration between MTA and Google will now transition to a full pilot project, with the Mountain View behemoth said to develop a production version that will be used by the track’s inspection team. If successful, this project can lead to considerable cost savings.
News Source: WIRED
About the author: Omar Sohail is a reporter and analyst for Wccftech's mobile section, specializing in the technology and business of the mobile industry. His expertise lies in the intricate hardware supply chain, covering developments in semiconductor manufacturing, chip lithography, and camera sensor technology.
Follow Wccftech on Google to get more of our news coverage in your feeds.
