Sidewalk Ada Design and Anomaly Condition Assessment Using Gopro Vibration Data

Sidewalks are critical components of urban transportation networks, facilitating short-distance travel and bridging the last-mile gap between destinations. The quality of sidewalks strongly influences pedestrians' travel experiences, particularly for individuals with mobility constraints. However, the extensive and complex networks of sidewalks pose challenges in inspection and maintenance, hindering their level of service in terms of travel mobility and safety. To facilitate the assessment of sidewalk design and condition more effectively, we propose a novel online approach for sidewalk design element detection and assessment, leveraging data collected from an inertial measurement unit (IMU) and a global positioning system (GPS) unit embedded within a GoPro 11 camera system. The algorithm identifies sidewalk surface defect locations by analyzing accelerometer data and classifying the data into five learned roughness levels, using online unsupervised learning. The proposed method is evaluated on a range of artificial obstacles with different pre-defined roughness and on a test route. The results indicate that this tool can significantly enhance both the accuracy and efficiency of the inspection process, facilitate large-scale sidewalk network assessments for use in accessibility analysis, and potentially save considerable labor and financial resources, especially when coupled with video-based reviews based on video footage collected from the same sensor.