Researchers have developed arrays of AI-controlled cameras and microphones to identify and monitor animals and birds in the wild – technology that they hope might help alleviate Britain's growing biodiversity issue.

Robot monitors were tested at three different places and gathered noises and photos that computers used to identify certain species and track their whereabouts. Dozens of different birds were recognized based on their songs, while AI analysis located and identified foxes, deer, hedgehogs, and bats with no humans to observe present. 

According to Anthony Dancer, a conservation specialist at the Zoological Society of London (ZSL) “The crucial point is the scale of the operation."

Dancer explained that tens of thousands of hours of audio from aforementioned sites and "identified all sorts of animals from them."

The specialist highlighted the importance of AI to make that happen. 

The project's test locations were chosen in London including Barnes, Twickenham, and Lewisham. The locations are gated off to prevent people from straying onto lines and are visited seldom by rail maintenance employees. They are owned by Network Rail, which played a vital role in establishing the initiative.

Dancer noted that since the crucial factor of starting the project has been established, he is hopeful it will allow for expansion into other areas.

Network Rail owns more than 52,000 hectares of land, and many of these places are critical to the conservation of the nation's biodiversity.

Neil Strong, biodiversity strategy manager for Network Rail spoke of bird species like the Eurasian blackcap, blackbird, and great tit and how AI was capable of detecting them from acoustic signals from sensors installed.

Strong believes it is "encouraging and provides important benchmarks for measuring biodiversity in the future.”

Six kinds of bats, including the common pipistrelle, were also detected by the AI monitors.

More hedgehog highways needed

Dancer explained to The Observer that “Bats almost certainly use railway bridges for roosting,” explaining that more detailed information can be obtained to pinpoint exact locations and protect them.

“In the past, we have had to estimate local wildlife populations from the dead animals – such as badgers – that have been left by the track or the roadside. This way we get a much better idea of population sizes.”

The project also revealed that other animals usually frequent rail lines including hedgehogs.

Strong voiced that usually, hedgehogs are constrained to certain locations "because they get fenced in," adding that there are possible solutions. 

Strong observes that "hedgehog highways" are springing up in Scotland, which include putting tiny holes into the bases of all new fences so hedgehogs can get in and no larger animals can get in. 

Network Rail and ZSL are working on expanding AI monitors in other areas like Chobham in Surrey and the New Forest.

According to Dancer,  “On the sites that we have already tested, we found signs of more than 30 species of bird and six species of bat, as well as foxes and hedgehogs, so we were pleasantly surprised with the relatively healthy levels of wildlife we found in London,” even though he notes that was not the main focus of the program.

The specialist remarked that the main goal was to see how AI technology could survey wildlife not only on Network Rail but also in other areas in the UK.

He expressed hope that the program could also provide insight into how species are responding to climate change and how vegetation should be managed on road verges and other areas, as well as allowing humans to protect biodiversity through a feat only AI can achieve- analyzing tens of thousands of hours of recordings and hundreds of thousands of images.