Animals colliding in traffic is a growing problem.
Over the past five years, there has been talk of collisions with deer such as elk, deer and elk by 50 percent.
In the past 10 years, the Norwegian Public Roads Administration has had various projects to do something about this problem.
They tried to scare the deer with electronic alerts.
– But this has had little effect, says Henrik Wildenschild of the Norwegian Public Roads Administration’s Traffic Safety Department.
Now engineering students at Høgskulen på Vestlandet have devised an alternative.
The solution the students see for themselves is to set up a camera in an area with lots of deer along the way.
As shown in the video, a red box will appear around the animal that appears on the camera.
The tech will then calculate the probability that what has been framed is actually a deer.
– When he is caught, a warning is sent, says Joachim Leros, one of the five students on the project.
The warning can be directed to the vehicle or on warning signs along the road.
It is difficult to distinguish a gazelle from others
A little over ten years ago, the Norwegian Public Roads Department had a project similar to what Western Norway University College students are currently working on.
But then the technology was not reliable enough.
There have been several reports that deer are not deer anyway.
Wildenschild now believes that students can participate in something.
If they can develop a system for capturing deer on camera, it can be combined with some of the projects we are working on.
The Norwegian Public Roads Administration is considering, among other things, introducing variable speed limits on certain sections with a lot of deer.
He points out that even if there is a good warning system in place, motorists must do their part.
He says many of them don’t slow down even if the signs warn that there may be animals on the way.
More accidents in the summer
A third of incidents involving deer antlers occur from June to August.
University lecturer Per Christian Engedal has our Student Supervisor.
It is believed that the project looks promising from the tests carried out so far.
“I think this is a technology that can bring more credibility to the notification that is given, so that people can adapt the speed to the conditions,” Engedal says.
He points out that some alarm systems today are based on models of reality, something that people can make Do not respond to alerts Because they do not trust the presence of gazelles in the area.
“Web specialist. Lifelong zombie maven. Coffee ninja. Hipster-friendly analyst.”