Google Pixel Phones Achieve 92% Success Rate in Subway Car Problem Detection; Potential Future Alternative to Costly Scanners

Since September of the previous year, a select group of Pixel smartphones have been actively deployed on four New York City subway cars, serving a critical role in safeguarding both commuters and operational efficiency. This initiative, a partnership between the Metropolitan Transportation Authority (MTA) and Google, aims to leverage cutting-edge technology to monitor and address potential issues within the subway cars, considerably reducing maintenance costs while enhancing safety.

Innovative Solutions: TrackInspect Technology in Action

Traditionally, human inspectors have been responsible for examining the extensive 665 miles of the New York City transit system, continuously on the lookout for problems like rail defects or water leaks. Additionally, specialized ‘train geometry cars’, equipped with a range of sensors, gather and transmit vital data regarding the condition of the infrastructure. However, Florida’s software engineering experts might be on the brink of a far more economical solution.

The Google Public Sector has introduced an experimental technology known as TrackInspect. When utilized with Pixel smartphones, this technology is capable of capturing essential audio, vibration, and location data. This information plays a pivotal role in training AI models designed to assist inspection teams in identifying and addressing potential issues proactively. During a testing phase lasting four months, Google’s devices successfully detected 92 percent of identified defect locations, which were subsequently validated by the inspection personnel.

MTA President Demetrius Crichlow envisions a future where a ‘modernized’ inspection system not only detects issues but also initiates repairs before they escalate into serious disruptions. According to Crichlow, this strategic approach is crucial for a transit system serving approximately 3.7 million daily commuters, significantly impacting the travel experience for countless individuals relying on the subway for their daily commutes.

The TrackInspect initiative has thus far amalgamated a staggering 335 million sensor readings and over 1, 200 hours of audio, utilizing New York City Transit’s existing database of track defects. This comprehensive data set has been instrumental in training approximately 200 specialized prediction models aimed at early issue detection. Following the promising results of this collaboration, MTA and Google are progressing towards a comprehensive pilot project. Google is expected to develop a production-ready version of this system for widespread use by the track inspection teams, paving the way for substantial cost savings and operational enhancements.

As these technologies continue to advance, the partnership between Google and the MTA represents a significant stride towards a smarter, more efficient public transit system, enhancing safety for millions and transforming the future of urban transportation.

News Source: WIRED

Source & Images