How AI Helps Save Lives On Roads

Europe is statistically one of the safest places in the world to drive your car on a road. Figures from recent data published shows that there are only a mere 50 fatalities per one million citizens. This is more than 3 times lower than the global average (excluding Europe) of 174 deaths per million citizens. However, we still have a long way to go to minimise the deaths from road accidents. In 2017 it was reported that approximately 25,300 people lost their lives while driving on European roads.

The cause of death is a result of many factors such as human error, bad weather, damaged roads and bridges, etc. Although human error cannot be directly fixed, the deaths due to unstable bridges and damaged road surfaces is something we can improve on which would directly lower the fatality rates. Although this may not be a result of initial design flaws, the problems stem from exposure to weather conditions which in turn erode the surface of the roads or cause metal on bridges to rust. More importantly, over time the damage will only get worse and if cracks go unnoticed then not only will it lead to increased repair costs but it also puts the life of civilians on roads at jeopardy.

This is where artificial intelligence comes in to play. Essentially, the goal is to create drones and cameras that are capable of interpreting and analysing any cracks or faults within these structures and then send a message back to report the location of the damage and its severity. Of course, this is done through advanced machine learning where data is fed to these artificially intelligent drones/cameras, making them capable of differentiating between smooth and cracked road surfaces.

AI drones capable of detecting cracks in roads and bridges.

AI drones capable of detecting cracks in roads and bridges.

Microsoft is partnering with Sund & Baelt and have developed a drone that utilises the technology mentioned above to help improve road safety. The drones fly around bridges, capturing thousands of pictures of the structures. This is much more efficient and safer than having workers traverse to different bridges and manually take pictures of any flaws or cracks. Instead, these workers are used to help train a machine learning algorithm which is then uploaded to Microsoft’s Azure cloud. Once the AI publishes locations and specific areas of bridges with vulnerabilities, workers are the deployed to these areas to maintain and repair the structures.

“Our core focus was to create a solution to maintain and improve efficiency, but we quickly realised that the more we used the solution, the better it actually became. This has given us an incentive to spread the solution to others,” states Mikkel Hemmingsen, CEO of Sund & Bælt.

With artificial intelligence being used in scenarios where the lives of people are at stake, it seems that the uses of AI extend beyond just businesses, phones, stock market predictions, etc. AI has the potential to be integrated in every scenario and we are slowly seeing this with many companies realising the benefits that AI brings with it.

Zacharia Sharif