Dynamic segmentation of smartphone sensor-based pavement functional condition

Abstract

Roads form a vital part of the infrastructure as it influences peoples’ lives directly in terms of connectivity as well as mobility. By using traditional methods for evaluating the pavement condition, the authorities spend a significant amount of resources. Automated techniques such as image processing and laser imaging systems can be adopted to overcome this issue. However, it needs complicated additions such as special lights and lasers that exponentially increases the surveying cost. In this explorative research, an effort was made to study the applicability and reliability of smartphone sensors in pavement functional condition monitoring. For efficient management of pavement for maintenance and rehabilitation activities, identifying the representative sections of these pavements would be more effective. A dynamic segmentation method is adopted to segment the smartphone-based pavement functional condition. The pavement was classified into five groups with 0–2, 2–4, 4–6, 6–10, and >10 m/km IRI values, as in Roughometer III.