May 28, 2020

Reducing risk in road projects with AI

The new route for the E39 between Mandalselva and Lyngdal passes through inaccessible and hilly terrain. Through EMerald Geomodelling, Nye Veier are using airborne geoscanning and AI to reduce risk in road development projects.

There are several risk factors in a road project. It is particularly challenging when the area to be investigated is inaccessible and hilly.

In early June you’ll see a low-flying helicopter with an attached geoscanning antenna between Mandalselva and Herdal in Lyngdal municipality.

Fewer surprises with geoscanning

Some information around subsurface conditions is available before scanning, but the verification of this and improved insights into ground conditions allows for easier road planning and construction. In particular it reduces the likelihood of surprises, such as the costly surprise of sulfide-containing rocks.

–  We hope not to find a strong signal, as this may indicate that there is in fact little sulfide-containing rocks in the area. We also hope to find out if there are any major weakness zones where tunnels are planned in the region, says Guro Skurdal, VP Operations at EMerald Geomodelling.

Acid-forming rocks require extraordinary measures when building infrastructure to prevent negative environmental effects, they similarly involve expensive costs in the treatment and handling of them.

Machine learning provides information in rough terrain

Road projects are prone to many risk factors, and are therefore heavily dependant on thorough preliminary investigations to make well informed decisions. This can prove particularly challenging in areas with inaccessible, undulating terrain.

–  It is important for us to make good road choices. We therefore want to know as much as possible, as early as possible. We have previously used EMerlad Geomodelling’s technology to successfully uncover ground conditions in two major development projects, says Nye Veier’s project manager Vidar Stormark.
–  Airborne geoscanning allows information to be gathered from difficult to reach topography. We are very pleased to be working in cooperation with Nye Veier, a company that dares to invest in new solutions and technology to achieve the best possible decisions, says Skurdal.

The geoscanning antenna which hangs below the helicopter collects data which is processed and with the help of machine learning, is used to generate a 3D model of the subsurface. The model is used to measure the geology in the area and makes it possible to plan the new road route at an early project phase. By combining existing data with that from the geoscanning, the machine learning algorithms provide good information over the area.

The signals emitted from the antenna are harmless for both humans and animals.

More case studies

All stories

Talk to us about your project

All we need to know is where your project is located, its size and particular geological unknowns. Based on that we can assess technological feasibility and economic benefit.


We scan the project area

Once we thoroughly assessed the added value of our method for your goals and received an order, you lean back and we care for data acquisitions, harvesting, integration and machine-learning based model building.


Work with an easy to understand 3D model

Working with EMerald you won’t need domain specialists to make sense of the sensed. We deliver and give easy access to integrated 3D models that provide the insights your project needs.

Let's talk about your project

Our team will get in touch with you shortly.

Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.