Pioneering Methods for Automated Curb Usage Pattern Recognition

Research Team: Murat Arcak (lead) and Alexander Kurzhanskiy

UC Campus(es): UC Berkeley

Problem Statement: A transformation is taking place in urban curb management: its stationary use as parking space is waning while it is increasingly becoming a space for dynamic interactions between people and vehicles. This transformation is fueled in part by the increasing public reliance on transportation network companies (TNCs), like Uber and Lyft, and on delivery services for food and consumer products. The curb space is also becoming an important location for bikeways, bus lanes, street vendors (e.g., food trucks), and paratransit pickups and drop-offs. How cities manage this public resource for the benefit of local economies is a timely problem, whose solution relies on first understanding emerging curb usage patterns over time and space. Currently no systematic method exists for identifying these patterns though there is great untapped potential for doing so with emerging machine learning technologies and non-traditional data sources, such as dashboard cameras mounted on shuttle buses, parking enforcement, and other vehicles that routinely traverse curb spaces. When combined with stationary cameras, these provide an excellent data source covering broad spans of time and multiple locations.

Project Description: To demonstrate how video data can be used to recognize usage patterns, the research team conducted a case study on Bancroft Way in Berkeley, CA. The research team collected video footage with GPS data from a dashboard camera installed on a shuttle bus that circles the area. The researchers trained a machine learning model to recognize different types of delivery vehicles in the data images, and then used the model to visualize curbside usage trends. The findings include identifying hot spots, analyzing arrival patterns by delivery vehicle type, detecting bus lane blockage, and assessing the impact of parking on traffic flow. The proof-of-concept study demonstrated that machine learning techniques, when coupled with affordable hardware like a dashboard camera, can reveal curb usage patterns. The data can be used to efficiently manage curb space, facilitate goods movement, improve traffic flow, and enhance safety.

Status: Completed

Budget: $80,000