Supporting road maintenance with in-vehicle data: Results from a field trial on road surface condition monitoring

Abstract

For road operators, road surface condition monitoring is a costly but necessary task. Usually, road operators have to conduct time-consuming and therefore costly inspection drives with specialized sensor vehicles gathering high precision measurements. This raises the question whether vehicle-data collected during daily routine drives could replace or at least supplement inspection drives. In this paper we present a prototypical implementation of a road surface condition monitoring system which analyses the vertical acceleration signal from vehicle-mounted accelerometer sensors (physical jerks). The work proposes a linearly referenced road surface condition index (RSCI), being derived from position and accelerometer measurements and being capable of describing vehicle-sensed road surface quality. A road operator supported field trial - using a prototypical system for raw data collection, analysis and integration of the derived road surface condition information into the road operator's legacy systems - was conducted to collect sample data as well as field experiences and to learn about the benefits of such an approach from the perspective of road operators. Results clearly demonstrate that the derived road surface condition information (RSCI, heavy jerks) is able to supplement existing maintenance information and provides a valuable additional data source for road operators by helping to focus monitoring and maintenance tasks. Beside the general applicability, several challenges and refinements on the way to a fully operational system have been identified.