Between September and January, six Google Pixel Smartphones have overcome free trips over four New York City Subway. Specifically, they took a train A, when it threw the ball 32 miles between the northern nose of Manhattan and the south of Queens.
The phones are not stoway or stray, and an extremely sharp passenger can be said because they are inside the plastic shell and are protected through brackets to the underside and interior of the car. Although people inside the car have used their smartphones to write emails or Instagram rolls or explore RoLlox, subway operators have used sensors of these phones, speed meters, from the next and gathering and for car -mounted people, external microphones added.
The phones are part of a short test of New York City Metropolitan Transportation Agency And Google Although the technology is cheap, mainly technology can complement the monitoring work of the agency. .Public area GoogleThe work -performing department has not charged MTA for this initial experiment.) Today, the tests were conducted by human inspectors, those who walked all 665 miles of the subway line of New York City, eyes peeled by problems such as broken rails, broken signals and water damage. Three -year trips by specialized, sensor geometric cars, also grasping and uploading more sophisticated data about the state of the city's railway infrastructure.
The work of New York City Transit with testing technology, which Google calls TrackinSpect, shows audio data, vibration and location, collected relatively cheap and used for training Artificial intelligence Predictable models can complement that test. It can guide people towards suspicious bells, bangs or hiss, suggesting the types of tools they will need to repair before they go there. During the four -month project, technology was able to identify 92 percent of the defect location and were then accurately defined by human monitoring inspectors, MTA said.
Finally, technology can become a way we can minimize the amount of work done to identify those errors and the inspectors on the right direction, so that they can spend time repairing instead of identifying and going straight into and doing work, Mr. Demetrius Crichlow, President of the agency said. In the future, MTA hopes to create a system of modernization online, automatically identify and organize corrections for monitoring issues.
For 3.7 million people's daily travel, catching defects before they becomes the problem may be the difference between working or going to school on time and being bogged down in unexpected delays.
The goal with this (project) is to find problems before they become a major service problem. Cooperation with Google will now expand into a fully pilot project, MTA said, where Google will build a production version of technology and put it in the hands of monitoring inspectors.
Utility of inspection
Brian Poston, Assistant Vice Chairman of Transport and Railway with WSP consulting firm, said that Google experiments are part of the bumper crop of the technology that supports AI that the new shipping agencies begin to use to supplement their typical tests, Brian Poston, a shipping and railway vice president with WSP consulting firm WSP. Although New York is the only use of Harmonics, the issues of vibration and vibration of accurate problems, others have installed small sensors or cameras on races with automatic measurements and flags difference when they appear. Technology is activated not only by advances in machine learning, but also cheaper and smaller battery and processor.
However, US regulatory agencies require regular railroad inspection and maintenance, and Poston says he does not expect those rules to disappear soon. Until the technology may be specific and accurate, you will always need that human interaction, he said.