Automatic Assessment of Dirt Roads Based on Crowdsensing

Unpaved roads are characterized by rapid deterioration due to erosive weather phenomena as well as the circulation of heavy vehicles. Poor road conditions imply significant costs that impact the productive activities in the surrounding area. Currently, the assessment of road conditions is conducted through visual inspection and, at times, digitization for georeferencing, which is considered a slow and costly process. In this work in progress, we present the development of a data acquisition system using commercial off-the-shelf (COTS) devices, such as mobile phones or embedded systems, to collect georeferenced and timestamped trajectory data. Building on the previously presented processing algorithms and simulations, the implementation of the data capture and tachograph data analysis scheme is detailed.