Experts knew wildfires that charred huge swaths of Australia in recent years killed hundreds of millions of animals, yet none had a firm idea of just how many died – or survived. Now an effort is underway to use drones backed up by artificial intelligence (AI) to make a more complete accounting of the nation’s wildlife.
Drones linked to AI computing hub will provide faster, more accurate wildlife population counts
A new network is being created around Queensland University of Technology’s (QUT) computing hub linking drones to IA platforms, thereby permitting a more thorough census of regional wildlife than can be achieved manually. The plan is to fly uncrewed aerial vehicles (UAV) equipped with infrared sensors to detect various life forms, which will then be identified by IA programs and machine learning capacities that expand the catalogue of recognized species as flights continue over time.
The program will be tested in the Noosa Heads area south of Brisbane on Queensland’s Sunshine Coast. That trial run will initially focus on counting koala populations, which were particularly hard hit by Australia’s 2019-2020 Black Summer fires.
QUT’s AI data hub will serve as the computing center of the program, while actual flights and other organizational work will be assumed by a variety of largely non-governmental organizations. Chief among those is Landcare, a vast grassroots network of individuals and collectives seeking to protect the environment, and pursue sustainable development projects. The nonprofit has donated $223,000 to the effort.
On-ground “game-changer” from the skies
The tech onboard drones, backed up by QUT’s AI capacities, is expected to speed the counting of wildlife, and provide more accurate tallies of animal populations by inspecting areas that have previously been difficult or impossible to access by human surveyors. UAV are also capable of scanning far wider areas of terrain in a single outing than people. Data transfers will also be accelerated via craft-to-computer exchanges.
“It’s a game changer,” Grant Hamilton, the head of QUT’s School of Biology and Environment Sciences and Centre for the Environment, told Noosa Today. “Despite decades of research, the 2019-2020 Black Summer bushfires highlighted just how little is known about the presence and abundance of wildlife across vast tracts of the Australian landscape.
“There were estimates of billions of animals killed or displaced – but these figures were by necessity based on modelling rather than monitoring, and this lack of data means priority areas for protection could not be established,” he continues. “As one example, the impacts of bushfire on koala populations in south-east Queensland were clearly evident but difficult to quantify because of a lack of baseline data.”
Previous animal surveying – often undertaken on foot – produced numbers whose accuracy Hamilton estimates at around 70%. Use of drones and AI computing to count wildlife populations, he notes, should initially average around 85%, and work toward 100% over time.
“This system will allow Landcare groups, conservation groups, organizations working on protecting and monitoring species to survey large areas in their regions, anywhere in Australia, with the use of drones and thermal imaging detection, and send the data back to us where we can process it,” Hamilton says. “It’s citizen science on a much bigger scale.”
FTC: DroneDJ is reader supported, we may earn income on affiliate links