What is it about?
This work introduces a way to identify mobility barriers for wheelchair users using a common smartphone mounted on a manual wheelchair. Traditional accessibility audits often miss dynamic problems like construction detours, rough terrain, or heavy doors. We log GPS, accelerometer, and gyroscope data along a university campus routes to provide real-time monitoring. We identify potential barriers by detecting anomalies in the sensor data. For example, acceleration spikes indicate issues like cracked pavement or curb transitions, while prolonged segments of near-zero acceleration signal a forced stop at a blocked entrance. Gyroscopic shifts highlight steep slopes or evasive maneuvers around crowds. We cross-reference these sensor signals with GPS data and annotated photographs to confirm the real-world accessibility barrier.
Featured Image
Photo by Erik Mclean on Unsplash
Why is it important?
This process provides actionable insights for facilities management and moves accessibility assessment beyond static compliance to a dynamic, inclusive system.
Perspectives
This research was highly motivated by the need to capture the genuine, day-to-day mobility challenges faced by wheelchair users that a simple static map can never reflect. Collaborating with the Disability Resource Center was crucial to ensure that our technical insights are grounded in the lived experiences of the community. I am excited about the future user studies, as they will refine our anomaly detection methods and help create a truly adaptive and inclusive monitoring system for accessibility.
Vivian Wong
University of Florida
Read the Original
This page is a summary of: Poster Abstract: Dynamic Mobility Barrier Detection Using Wheelchair-mounted Sensors, November 2025, ACM (Association for Computing Machinery),
DOI: 10.1145/3736425.3772111.
You can read the full text:
Contributors
The following have contributed to this page
