A while ago, we posted about our application to the Transit Tech Lab, an accelerator run as a public-private partnership between the Metropolitan Transportation Authority (“MTA”) and the Partnership for New York City whose mission is to apply emerging technologies to public transportation problems in the city’s bus and subway systems.
Our team at Palisade Labs is excited to announce that we’ve been selected as one of the program’s six finalists!
Over the next two months, we will be working with the MTA and domain experts to explore the use of computer vision for delivering tangible insights on bus lane obstructions and general bus delays. Using video footage from a bus’s forward-facing dash camera, we’ll be tackling questions such as how a bus spends its time (moving, in traffic, picking up passengers, etc.), how often a bus is obstructed, and what types of obstructions occur. The resulting insights would be acted upon for traffic enforcement or urban planning purposes to improve bus speeds and efficiency.
With over 5,700 vehicles, 320 routes, and 760 million annual ridership, the MTA bus system carries more passengers than the next three largest domestic systems combined. For many New York residents, especially those in outer-borough, low-income neighborhoods underserved by subways, public buses are often the only affordable transportation option, with many solely depending on these services for commuting to work, school, and appointments.
Bus travel speeds are limited by traffic congestion in city streets. In midtown Manhattan, the average bus speed is as low as 3.9 mph, comparable to an average New Yorker’s walking speed. Speeds are better in the outer boroughs, but not by too much – Brooklyn’s buses average around 7 mph. Using computer vision to sift through volumes of bus video footage could be key to understanding root causes of bus delays, a first step to modernizing and increasing public bus efficiency.
When it comes to working with the MTA, it’s hard for us to take our New Yorker hats off. New York City is thoroughly embedded in the identity of our company. In fact, our company name derives from the New Jersey Palisades, a line of dramatic, basalt cliffs just on the other side of the Hudson River. The company’s partners, Jacob, Dylan, and I, have spent almost all our professional lives working at and forming companies in Manhattan and Brooklyn. Additionally, backing our team is a deep bench of informal advisors all with ties back to the City. We view it as our civic duty and an honor to work with an institution as storied as the MTA, with its rich legacy and vital importance as the lifeline of New York City. (We also personally just can’t wait to help the buses in our neighborhoods run faster!)
As we embark on this project, we’ll be blogging about the technical challenges we encounter, the models we use and develop, and any insights we uncover. We look forward to engaging the computer vision and data science communities as we stand upon the shoulders of giants in utilizing existing tools and data sets that many have poured countless hours into developing. Please stay tuned for further updates!