The New York City Subway Is Using Google Pixels to Listen for Track Defects

With September January six Google Pixel Free walks of smartphones New York City Metro cars. In particular, they took a train like pinging 32 miles between the southern ends of the queens with the northern end of Manhattan.

Phones could not be in stowaways or streets and for the extremely sharp-eyed passenger plastic coatings and to be provided with parentheses in parentheses. In order to write emails in people inside cars or to explore the roblox, subway operators were used for the sensorometers of these phones, magnets and giroscopes, additional external microphones.

Phones were part of a short experience of New York City Metropolitan Transport Organization and Google Cheap, mainly, the agency can provide track inspection work in outside technology. (Google Public SectorToday’s the case has not filled the MTA for this initial experience.) Today, today, all 665 miles of metro roads are carried out by human inspectors such as all 665 miles of subway roads, sound signals and water damage to human inspectors. Special, Sensor-Laden “train geometry cars” by three years of walking, the city’s railway infrastructure also draws more complex information and loads.

New York City Transit offers sound, vibration and location data used to work with Google Trustinspect’s application technology, relatively inexpensive and exercise artificial intelligence Forecast models, this can add this verification work. People can direct suspicious noises, bangs or leaps, bangs or squeals can suggest what tools they need to do before repairing before repairs. Throughout the four-month project, technology was later able to identify 92 percent of the defect, which was determined by human track inspectors.

Finally, technology “can minimize the amount of work done to identify these defects and the amount of work done to indicate the right direction, so it can spend time adjustments and work instead of determining,” said Demetrius Crichlow. In the future, the MTA hopes to create a “modernized modernized” system that automatically identifies the adjustments for tracking problems.

Before there are problems for the system 3.7 million daily drivers, there may be a difference between capturing, timely or school and fall into school.

“By the purpose of [project] “It is to find problems before being a great issue,” said Crichlow. Cooperation with Google will now expand to the full pilot project.

Gadgets of inspectors

Google experience, transit agencies are part of a bumper product of an effective technology, which began to use to use transit institutions. New York is unique to use “Harmonics” and the vibrating points, others, when they emerge, they install small sensors or cameras on roads that take automated measurements and inconsistencies. Tech is not only in the machine learning, but also in cheaper and smaller batteries and progressors in processors.

Again, US regulators regularly require human inspection and rails, and Poston does not expect these rules to go any time soon. “It will always need human interaction until technology can be specific and accurate,” he says.