Researchers have created arrays of AI-controlled cameras and microphones designed to detect and monitor animals and birds in their natural habitats, offering a potential solution to the growing biodiversity problem in the UK.
These robotic monitoring systems have undergone testing at three different locations, capturing both audio and visual data. AI-powered computers were able to use this data to identify specific species and track their movements. The technology successfully recognized a variety of bird species based on their calls and also identified animals such as foxes, deer, hedgehogs, and bats. Notably, these identifications were made without any human intervention.
Anthony Dancer, a conservation specialist at the Zoological Society of London (ZSL), emphasized the significance of the project’s scalability. He mentioned, “We have gathered a vast amount of data files and hours of audio from these test sites, leading to the identification of various animals. Achieving this using human observers would have been impractical; AI made it achievable.”
The chosen testing sites were situated alongside railway lines in London, specifically at Barnes, Twickenham, and Lewisham. These areas, owned by Network Rail, were selected due to their relatively undisturbed nature, facilitating the initiation of the project. The successful outcomes at these sites have paved the way for the expansion of the technology to other regions.
Network Rail holds ownership of over 52,000 hectares of land, which plays a pivotal role in protecting the country’s biodiversity. Neil Strong, biodiversity strategy manager for Network Rail, underscored the technology’s relevance in identifying species like the Eurasian blackcap, blackbird, and great tit. The AI system was able to detect these bird species through the acoustic signals captured by sensors at the test sites, providing essential data for future biodiversity assessments.
AI-powered monitors also identified various bat species, including the common pipistrelle, which often use railway bridges as roosting locations. Enhancing the understanding of roosting locations through AI monitoring could significantly contribute to conservation efforts.
Future plans involve expanding the implementation of AI monitoring to additional regions such as Chobham in Surrey and the New Forest. The application of this technology goes beyond railway environments, informing conservation strategies across diverse habitats.
Dancer emphasized the broader objective of the project, stating, “AI-driven technology, combined with acoustic and camera traps, can effectively survey wildlife not only on Network Rail land but also in other areas of the UK. It will help us understand how species respond to climate change and guide vegetation management, not only along rail lines but also in roadside areas and various other locations.”
Given the challenges posed by climate change to biodiversity, the role of AI, particularly in machine learning, becomes pivotal in analyzing the extensive data required for effective conservation efforts.
