SCC-PG: Democratizing AI-based Urban Mobility Data Acquisition for Underserved Communities by Battery-powered Smart Cameras

Effective urban planning and transportation system designs require comprehensive urban mobility data, such as pedestrian statistics. However, existing data collection methods, which rely on manual counting or AI-based video analysis from fixed traffic cameras, often overlook underserved communities such as older adults who need safer and more accessible public spaces to navigate their day-to-day activities.

This project aims to address these data unfairness and scarcity issues by developing and deploying low-cost, battery-powered smart cameras. These devices will enhance the fairness and scalability of data acquisition, particularly benefiting older adults, people with disabilities, and other underserved communities in accessing mobility services. By ensuring privacy-preserving data acquisition, the project aligns with the NSF's mission to promote the progress of science and advance national health, prosperity, and welfare.

This initiative will contribute to more equitable urban planning and foster significant advances in smart city technologies.

The proposed project introduces an innovative effort to democratize urban mobility data acquisition by developing and deploying low-cost, battery-powered smart cameras. The research will focus on creating new control and perception algorithms to operate these devices efficiently within the constraints of battery life and computational capacity. Specifically, the project will develop (1) on-device adaptive sampling techniques to optimize power consumption by enabling smart cameras to perceive and react to real-time environmental changes, and (2) methods for uncertainty quantification in AI-based video analysis outputs to support robust and resilient decision-making in dynamic urban settings.

The project includes prototyping and pilot field experiments in underserved senior communities in NYC. Collaboration with Geriatrics researchers will explore the social aspects of this technology, enhancing our understanding of how underserved communities perceive and accept AI technologies in transportation services. This interdisciplinary research will bridge AI, transportation, computer vision, and Geriatrics, promoting data fairness and energy-efficient AI.

The project's open-source hardware design, algorithms, and privacy-preserving data will facilitate broader applications and contribute to senior safety and healthy aging.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.