Unmanned aerial vehicles (UAVs), also known as drones, don’t offer just a safer way for wildlife biologists to observe their subjects; they’re often less costly, more efficient and more precise than traditional approaches. While researchers are still navigating the challenges and complexities of mixing wildlife and drones, the technology holds incredible potential. In fact, drones are already changing the way that data is collected.
In the past few years, these flying robots have started to provide an unprecedented look at animals and their hard-to-reach habitats—such as orangutan nests high in the jungles of Borneo—and have shown potential for catching poachers and stopping illegal logging. And because they carry digital cameras that produce geo-referenced photos, the data they gather can be fed into image-recognition algorithms to vastly improve the accuracy of population counts.
While biologists are incorporating more and more UAVs into their fieldwork, drone scientists are working to improve the autonomy, endurance and maneuverability of drones. Two recent developments are noteworthy: 1) researchers have just discovered that birds fly more efficiently by folding their wings during the upstroke, and this could mean that wing-folding is the next step in increasing the aerodynamic and propulsive efficiency of flapping drones; and 2) a new technique using imaging drones and artificial intelligence is allowing researchers to study the behaviors and movements of group-living animals in a 3-D structure of their environments.
Flapping drone wings
Among the flying animals that are alive today, birds are the largest and most efficient. Engineers and scientists find that this makes them particularly interesting as inspiration for the development of drones. However, birds have different flapping strategies that optimize the energy costs associated with the duration and speed of their flights, and determining which one is best for a drone requires aerodynamic studies of the various ways birds flap their wings. For example, as the precursors to birds—extinct, birdlike dinosaurs—developed active flight, they benefited from folding their wings during the upstroke.
So, to determine which flapping strategy would work best for a drone, a Swedish-Swiss research team recently constructed a robotic wing that can flap more like a bird’s wing than previous robots could but that also waves in ways that a bird’s wing cannot. By measuring the performance of such wing movements in a wind tunnel, the scientists were able to observe how different methods of achieving the wing upstroke affect energy and force in flight.
Previous studies had shown that birds flap their wings more horizontally when flying slowly. This new study, published in the science journal Advanced Intelligent Systems in December 2022, shows that even though it requires more energy, birds probably do it because it makes it easier to create a sufficiently large force to stay aloft and propel themselves. This is something drones can emulate to increase the range of their speeds.
Because research into the flight abilities of living birds is limited to the flapping movements that birds actually use, the scientists state that their new robotic wing can be used to answer questions about bird flight that would be impossible to understand simply by observing flying birds.
By helping to explain why birds flap the way they do, and by finding out which movement patterns create the most force and are the most efficient, these results will also add to our understanding of how the migration of birds is affected by access to food and climate change.
An additional area where these insights could be put to good use is when using drones to deliver goods. Flapping drones would need to be capable and efficient enough to lift the extra weight that deliveries entail, and wing movement is of great importance for this kind of performance.
In situ group dynamics
But better comprehension of bird migrations and more efficient food deliveries aren’t the only areas of drone research that are being advanced. Imagining drones are becoming adept at studying entire groups of animals in their natural habitats.
To explore groups of animals in their native environments, such as a herd of plains zebras in Kenya or gelada monkeys in Ethiopia, researchers from Aarhus University in Demark and the Max Planck Institute of Animal Behavior and the Center for the Advanced Study of Collective Behavior at the University of Konstanz in Germany developed a new method for collecting data about the behaviors of animals and their surrounding natural landscapes using computer vision and drones.
Plains zebras are interesting for collective and spatial behavior studies, say the researchers, because they live in multilevel societies: small groups of females and a male combine to form larger herds of dozens of animals. This social and spatial structure could influence behavioral processes, such as decision-making and information-sharing—and have implications for understanding our own complex societies.
In the past, researchers got precision data about animal group dynamics mostly in highly controlled lab conditions where it was possible to repeat experiments over and over. But, wondered the Denmark and Germany team, could new computer algorithms and imagining drones be used to bring the same lab approaches into natural landscapes?
A couple of challenges had to be overcome: first, the scientists were often recording 20 or more individual animals at a time. Quantifying where each of them was in a single, half-hour video could take weeks. So, the first hurdle was figuring out how the animals the researchers were interested in could be automatically detected. The solution was training powerful computers to do some deep-learning algorithms.
Second, the researchers were interested in individual animal movements, yet the videos they were able to make included not only wildlife movements but also camera movements and distortions from the hilly landscapes where they were filming.
However, since drones not only observe an animal group but also the landscape that it is situated on, a very broad dataset is obtained, which includes information on the environmental and social context of all the animals in the group. This is possible because drones model the 3D landscape they are recording. That allows researchers to examine the effects of habitat on behavior, a powerful approach that had previously been very difficult.
Another advantage is a drone’s unobtrusiveness, say the researchers, who published their results in the Journal of Animal Ecology in March 2023. New drones fly so high overhead that wildlife isn’t bothered by them. And, unlike another common method, animals don’t need to be captured and fitted with movement sensors, which can be an expensive and risky procedure, especially when working with endangered species, such as Grevy’s zebras.
Worldwide, wildlife populations are declining due to climate change, habitat loss and other threats. Learning more about how groups of animals behave in complex, natural environments can help us to create more effective conservation actions and generate new insights into the behaviors and lives of wildlife species.
Drones certainly have their drawbacks, and they have caused significant problems in nature. But they’ve also been used to benefit the natural world and to give us passage into the secret lives of wildlife. Developing more efficient ones—with designs inspired by our fellow creatures—could help us better protect our planet and its precious biodiversity.
Here’s to finding your true places and natural habitats,
P.S. Watch the video below, which demonstrates how a new drone captures information from a herd and then transfers it to a computer for further processing. The first part of the animation is designed to look like a small exhibit, with a diorama of the Kenyan landscape. The second part illustrates how a computer handles the data given to it. From the individual animal and landscape information picked up by GPS, scientists can then reconstruct how an individual in the herd moves through the terrain (symbolized by the abstract landscape you see at the end) and interacts with its group.