Connected Vehicle Approach for Pavement Roughness Evaluation

Abstract

Connected vehicles present an opportunity to monitor pavement condition continuously by analyzing data from vehicle-integrated position sensors and accelerometers. The current practice of characterizing and reporting ride quality is to compute the international roughness index (IRI) from elevation profile or bumpiness measurements. However, the IRI is defined only for a reference speed of 80 km/h. Furthermore, the relatively high cost for calibrated instruments and specialized expertise needed to produce the IRI limit its potential for widespread use in a connected vehicle environment. This research introduces the road impact factor (RIF), which is derived from vehicle integrated accelerometer data. The analysis demonstrates that RIF and IRI are directly proportional. Simultaneous data collection with a laser-based inertial profiler validates this relationship. A linear combination of the RIF from different speed bands produces a time-wavelength-intensity-transform (TWIT) that, unlike the IRI, is wavelength-unbiased. Consequently, the TWIT enables low-cost, network-wide, and repeatable performance measures at any speed. It can extend models that currently use IRI data by calibrating them with a constant of proportionality.