AI camera from Flox Intelligence to deter wildlife from railway tracks

Combating wildlife accidents with Artificial Intelligence (AI): An AI-based detection system is intended to recognize wildlife at an early stage and help to deter it from the railway tracks. This is intended to benefit rail transport and nature.

At the end of 2024, Flox Intelligence and the French railway vehicle manufacturer Alstom began a field test in which the AI is supposed to recognize various animal species near the tracks. In April, the second phase of the field test began, which uses acoustic deterrence in addition to video recognition, as the two companies announced. The tests are being carried out on several routes of the Swedish railway company Tåg i Bergslagen.

The system comes from the Swedish company Flox Intelligence. It has developed a camera that recognizes wildlife using AI. The cameras are set up along the railway line and are intended to scare animals away with acoustic signals when they approach the tracks.

The system learns

The system recognizes many different animal species, from birds to elk, deer, foxes, or wild boar. It categorizes each detection and is thus intended to improve its capabilities continuously.

The AI is particularly good at recognizing domestic animals and birds such as crows and pigeons. Elk and deer, on the other hand, required more training effort to achieve the same level of accuracy. "Our field tests show that AI has the potential to identify animals with high precision," said Flox Intelligence CEO Sara Nozkova.

In Sweden, there are around 5,000 collisions with animals every year. These harm nature as well as traffic: animals are injured or die, there are delays in rail transport, trains are damaged and need to be repaired. In addition, such accidents also have emotional effects on train drivers.

"We were positively surprised at how many animals we could observe from just a single train, on sections that had many collisions," said Maria Signal Martebo, Managing Director of Alstom Sweden. "The tests so far have given us a better understanding of which species move near the tracks, how effective the existing wildlife fences are, and how the technology can contribute to both safer transport and new knowledge about wildlife along the railway."

(wpl)