Coleta e análise de dados provenientes de acelerômetros de smartphone durante o trajeto de um veículo visando detecção de descontinuidades em vias urbanas

Abstract

Transportation plays a crucial role in the functioning of society, both for individual and collective mobility in urban areas and for the flow of goods that sustain various industrial and commercial sectors. The quality of roadways directly affects vehicle preservation, traffic flow, and driver experience. In this context, monitoring pavement conditions becomes a valuable tool for road maintenance. This study proposes the use of smartphone accelerometers to detect pavement anomalies as a means to support the upkeep of road infrastructure. Data collection was conducted on selected road segments in the city of Monte Alto – SP, and subsequent analyses were carried out to identify surface irregularities. The objective was to detect and classify discontinuities present in urban pavement – such as speed bumps, raised crosswalks, road studs, and potholes – through the development of classifiers using time-domain metrics. The data acquisition process and the mathematical methodology employed proved effective, yielding promising results with an F1-score of up to 98.7% in pothole detection.