Photo: Innovationsfonden

Cars to report on road condition

Mapping and surveying Data analysis Sensors Mathematical modelling Mobility
Leveraging modern car technology to continuously collect road data may help produce real-time digital images of road maintenance needs.

This could mean savings as well as increased road user safety.

Modern cars are equipped with a wide range of sensors that collect a variety of data while driving. These data can be directly or indirectly related to the condition of the specific road on which the car is driving.

Collecting and processing these car data creates a brand new opportunity for exploiting previously unused data sources for making ongoing and automated assessments of the condition of the Danish road network, which is the main purpose of the new LiRA (Live Road Assessment) project, with DTU Civil Engineering and DTU Compute as contributors.

The project, which is funded by Innovation Fund Denmark with EUR 1.6 million (DKK 12 million), will make it possible to collect data from the sensors of ordinary passenger cars and develop related models for assessing road condition. Road wear and tear can thus be discovered much earlier and road maintenance will therefore be much more efficient—requiring fewer resources.

Asmus Skar, a postdoc at DTU Civil Engineering, is spearheading the project’s data collection efforts.

“The goal is to exploit the recent technological development of passenger cars, which has meant that all cars are now equipped with a wide range of sensors with the potential for collecting data while on the road. We are researching what it will take to collect the data and are developing a model that can potentially digitize the condition of the road network, so you can quickly locate the areas in need of maintenance and improvement,” he says.

One of the challenges is ensuring that the data collection takes into account—and compensates for—the fact that we all drive differently and that the weather is changeable, which may also affect the data collected.

Using machine learning, a comprehensive mathematical model enabling assessment of the physical condition of roads will be developed in collaboration with colleagues from DTU Compute. This could revolutionize road infrastructure maintenance and thus ensure considerable savings.

Revolutionizing road assessment
Among other things, the model will be able to identify road damage that is developing faster than conventional road assessments can be carried out. Furthermore, it will be able to supply data on road conditions during the winter season where conventional measurements are not carried out due to the adverse weather conditions.

“The ambition is for the innovation of this project to help revolutionize condition assessments of the Danish roads. In addition, the new project will enable analyses of noise and carbon emissions, which can be used as input for the maintenance strategies and contribute to socioeconomic calculations,” says Matteo Pettinari, who is a specialist at the Danish Road Directorate and Project Manager at LiRA.