Artificial Intelligence can assist in daily road maintenance

RoadAI
Markus Melander, Head of Business Development, CV Solutions
Markus Melander
Head of Business Development, CV Solutions
Published:
Roads
Weather & Environment

Everyone that has been in a survey vehicle doing road inspection, or has served as road maintenance crew, is aware of the complexity and fast decision-making needed on the road. Sometimes it may be too risky to slow down to analyze different situations in detail or you may miss some maintenance needs while trying to do many things at once. RoadAI has been developed to address challenges such as these. More specifically, RoadAI can help to provide safety and simplicity for inspectors while they make assessments and take notes. Also, computer vision algorithms can automate some of the road network data inventories. 

The Vaisala RoadAI system combines a road video system and state-of-the-art analytics for road inspection in terms of pavement condition, asset management, road works management, driving conditions, weather conditions, and so on. The combination of road videos, video notes, and artificial intelligence supporting the road inspector is highly valuable in this challenging environment.
Basically, the RoadAI system enables road inspectors to have an artificial assistant available to support them in their work. 

Users can get the best value out of the system when they are able to combine all the different benefits into the same process and improve road management with up-to-date road videos and data generated by software algorithms. A good example is when road maintenance crews can collect road view videos and create notes or service tickets through speech as they drive. Data is shared to other users and image data on the map supports managers in their decision-making. RoadAI can also provide good proof and visual validation if there are claims or issues related to road maintenance. 

For planning, road videos provide excellent support for identification and prioritization to define actions needed. Because videos and virtual access to the network is available online, all parties across the organization benefit from the system.

How does RoadAI work?

RoadAI automated analysis and mapping is based on computer vision algorithms trained together with pavement distress specialists and trained road inspectors. The system is simple by nature. Recorded video data is processed with artificial intelligence. The system detects defects or other defined parameters from the pavement or infrastructure, and also maps and identifies road assets. Detection results are provided in a map interface to be used in work planning process. RoadAI pavement analysis can be used also for seasonal road inventories. This novel solution does not require a specialized vehicle as videos can be recorded from any car. 

In discussions with maintenance managers, one of the most often heard arguments on behalf of RoadAI is that it is so convenient to be able to see the network and make notes without even leaving the office. Or that they value taking notes without taking their hands off of the steering wheel. In general, it is useful for an organization when they can collect quality data on road conditions, plan investments, and react to problems before they get worse and become a risk for traffic. 

As for data privacy, the system is safe to use as all data is anonymized and there is little possibility to have humans or vehicles captured into the field of the vision.