ABSTRACT
With the advance of sensing technology, 3D laser imaging has gained much attention for its ability to accurately measure 3D pavement surface with 100% coverage both transversely (more than 4,000 data points per transverse profile) and longitudinally (up to 1 profile every 1 mm along driving direction). It is now possible to derive complete 3D pavement characteristics that, according to the literature, can reflect the possible causes of rutting but have been previously difficult to obtain at a large scale. The objective of this study is to develop a method for categorizing pavement rutting by the causes. Using 5-year 3D sensing data collected on pavement sections with comprehensive rutting and traffic conditions, the proposed method first derives the following features of 3D rut shapes: (1) Spatial features, including characteristics of transverse profiles (e.g., rut depth, rut width, and cross-sectional area) and longitudinal characteristics, such as rut length; and (2) Temporal features, such as the change and rate of change of the elevation and cross-sectional area. Using 67% of the data for training and 33% of the data for testing, a supervised SVM classification model is constructed to categorize 3D rut shapes into groups that represent different causes of rutting. Results of this study indicate that the proposed method provides a new means for monitoring and assessing pavement performance, which can potentially help transportation agencies more accurately and effectively identify cause of ruts to support data-driven pavement management decision-making.
