How Can AI Improve Our Future Road Safety?

Road safety is one of today’s foremost concerns, and Kawasaki Geological Engineering found a solution in Fujitsu’s technology. The company has been performing geological surveys in Japan for over seventy-five years. It helps the civic bodies discover road cavities due to ‘pipework, soil erosion or seismic activity’. So, how does artificial intelligence help in road safety? Let’s find out in this article at I-CIO by Amit Roy Choudhury.

Road Safety and AI

According to KGE specialist engineer Toshimune Imai, sinkholes resulted in more than 3,000 accidents in Japan alone, a bottleneck for fast-paced growth. So, KGE has designed radar tools to send electromagnetic waves for up to 3 meters into the ground. They pick up massive volumes of data as they move along on the road surface at 60 kph. A group of skilled technicians goes through the information together to prevent oversight. However, this way of finding cavities for road safety seems to take up a lot of time and effort.

KGE partnered up with Fujitsu’s Traffic & Road Data Service unit to use their Zinrai Deep Learning platform and build an AI technology. They trained the innovative tools to distinguish between sinkholes and other underground objects before leveraging them.

The Road to Refinement

Imai confirmed that they wanted to restrict the AI training to determining road cavities. They soon found out that they need to broaden their training framework so that the tools can differentiate a pebble from a crater in the ground. The machines, equipped with sufficient data, can discover issues in a 2-km road in a minute while three engineers would take an hour. So, the new technology enabled KGE to reduce effort by 90 percent.

Imai observed that humans narrow down their goals and pursue that in on a broader data collection. In contrast, a machine can objectively analyze all data. KGE aims to build a ‘compact road survey unit’ with better mobility and does not require highly experienced technicians for supervision. They plan to install these devices on generic patrol vehicles to send data for analysis sooner to address road safety concerns.

To view the original article in full, visit the following link:

Show More
Back to top button

We use cookies on our website

We use cookies to give you the best user experience. Please confirm, if you accept our tracking cookies. You can also decline the tracking, so you can continue to visit our website without any data sent to third party services.