MTA Leverages Google Pixel Technology to Revolutionize Subway Track Inspections and Enhance Safety

Anyone familiar with the New York City subway knows it faces numerous challenges, ranging from unusual sounds to potentially hazardous materials on the tracks. In an attempt to address these issues, the Metropolitan Transportation Authority (MTA) is currently experimenting with how AI might enhance the repair workflows using six Google Pixel smartphones.

In this particular initiative, the Google Pixel devices traveled across four distinct subway trains from last September to January. This investigation, conducted in collaboration with Google Public Sector, utilized the smartphones’ accelerometers, magnetometers, and microphones to detect any alarming sounds. The collected information was subsequently transmitted to cloud systems, which utilized machine learning algorithms to produce predictive insights.

This technology, referred to by Google as TrackInspect, was able to identify 92 percent of defect sites located by human inspectors. «The ability to detect early rail defects not only saves money but also conserves time — for both maintenance teams and passengers,» stated New York City Transit President Demetrius Crichlow in a statement. «This pioneering initiative — a first of its kind — employs AI technology to enhance the riding experience for customers while also making the work of track inspectors safer by providing them with more sophisticated tools.»

Traditionally, inspectors are required to traverse all 665 miles of subway tracks to identify any problems, complemented by sensor-equipped “train geometry cars” that collect data three times annually. During this trial, inspectors evaluated any areas flagged and verified the presence of defects. Additionally, they could pose inquiries regarding maintenance and protocols through the generative AI system.