Demystifying the Mysteries of Elephant Trunk Biomechanics

MathWorks Editor
MathWorks
Published in
8 min readNov 8, 2021

Discoveries Could Spur Biomimicry in Soft Robotic Design

Kelly, a 34-year-old African elephant at Zoo Atlanta, had never encountered corn chips before. For her first interaction with one, she used suction to levitate it deftly into her trunk’s grip. The fragile tortilla chip stayed completely intact. She awed the zookeepers and mechanical engineers from the Georgia Institute of Technology who were looking on by successfully repeating this delicate maneuver with her powerful trunk several more times.

“It almost feels like David and Goliath,” said Andrew Schulz, a Ph.D. student at Georgia Tech in professor David Hu’s Laboratory for Biolocomotion and first author on the study. “This should not be happening.”

The researchers already knew that elephant trunks are muscular appendages weighing over 100 kilograms (220 pounds) and used for breathing, sound production, touch, olfaction, and manipulation. But Kelly’s chip levitation was a surprise. Working closely with Zoo Atlanta keepers to design the study, they could see exactly what was happening, allowing them to shed light on an area of biomechanics that’s long been a mystery.

Georgia Tech engineers studying elephant trunk biomechanics envision their research advancing biomimicry, inspiring new soft robot designs, and strengthening conservation efforts to protect elephants in the wild.

Kelly regulated her trunk and took the chip from a table-sized force plate using suction to get it into the two “fingers” on the tip. The engineers imported data from the force plate directly into MATLAB® while they were at the zoo and produced visualizations on the spot.

“To our amazement, we saw the force was nearly zero,” Schulz said. Kelly had regulated her trunk to use less than 5 newtons of force on the chip. High-speed video revealed her two nostrils dilating, which led to more careful experiments, including presenting her with rutabaga cubes and capturing ultrasonic imaging of her trunk while she drank water mixed with chia seeds.

Schulz, Hu, and their team published their groundbreaking findings in the Journal of the Royal Society Interface. Flow visualization contributed to a mathematical model showing that Kelly’s nostril dilation increased her nasal volume by 64%. The study also demonstrated that elephants use their lungs and trunks to inhale at speeds over 150 meters (nearly 500 feet) per second.

The engineers envision these impressive capabilities advancing biomimicry, inspiring new soft robot designs, and strengthening conservation efforts to protect elephants in the wild.

Exploring New Territory

The gaps in scientific knowledge about elephant trunk biomechanics are more like chasms. In the 1870s, English photographer Eadweard Muybridge’s series of black-and-white images captured a horse in motion, finally answering questions about the animal’s physiology. Elephants, however, would have to wait much longer.

“The most detailed anatomy of an elephant trunk was done in 1908. It’s a hand drawing,” Schulz said, referring to the monograph of an Asian elephant by Johan Erik Vesti Boas and Simon Paulli. Prior to the latest Georgia Tech research, the most recent biomechanics study of elephants was completed four decades ago.

One of the main reasons for this incredible lag is elephants’ massive size. A 31-year-old male African elephant currently at Zoo Atlanta weighs more than 4,500 kg (10,000 pounds). Scientists must devise scientific experiments on an extremely large scale. The animals are incredibly powerful and frequently break the experimental setups, Schulz said. He added that tongues and octopus arms, similar to trunks, tend to be more accessible.

“A lot of this can be challenging because when I have a question that’s not answered, normally no one in the world has answered it,” Schulz said.

1908 monograph of Asian elephant by Johan Erik Vesti Boas and Simon Paulli. (Image credit: Biodiversity Heritage Library)

Elephant research is challenging due to the animals’ massive size, power, and tendency to break experimental setups.

Any study has to be safe for the elephant, provide stimulus, offer enrichment, and advance conservation. The behavior involved shouldn’t be taxing for them. “In looking at what they’re drinking, eating, and breathing, it’s important that all of our setups only have things that are natural to them,” he said. “We can’t use treated wood because their trunks are not used to those kinds of chemicals. We have to be very careful.”

Getting an Insider’s View

Schulz spent several months volunteering as an intern at Zoo Atlanta, mainly cleaning up after the elephants and doing some behavior and enrichment activities with them under the keepers’ guidance. Elephants are foragers, so he spread out browse — the food they eat every day — for them.

These firsthand experiences directly influenced the experimental setups. Schulz learned that African elephants usually eat a great deal of taro root in the wild. When his team considered what to use for the study, they selected rutabaga, a root vegetable with a texture similar to taro, over other possibilities like carrots.

“It was the first time we could see what’s happening internally to the trunk.” - Andrew Schulz, Ph.D. student at Georgia Tech

“With rutabaga, it’s easy to have larger or smaller cubes so we can regulate the graph of X to Y, where we’re looking at different sizes and masses,” he said. They performed 42 experiments using rutabaga cubes cut with side lengths of 16, 32, and 64 millimeters (0.63, 1.26, and 2.52 inches). The lower the mass and the greater the food size, the more likely Kelly was to use suction on it.

The idea to mix 50 grams (1.7 ounces) of chia seeds in water the night before the siphoning experiment came from a zookeeper. “A lot of times you use flow visualization with fluorescent beads, but you can’t do that with an elephant,” Schulz recalled. “Using chia seeds was a really good way for us to get flow visualization. You can even see the chia seeds in the middle of the trunk accelerating faster than the ones on the outside.”

Veterinarians regularly check health inside elephant trunks using ultrasonographic imaging. This approach opened up new research possibilities for the Georgia Tech team. Schulz inquired whether they could obtain ultrasound data from the trunk while Kelly snarfed water thickened with chia seeds. The answer was yes, so the vet applied gel and then placed the probe on the elephant’s skin during the experiment.

“I don’t think anything like that had ever been done before,” Schulz said. “It was the first time we could see what’s happening internally to the trunk.”

The researchers used MATLAB for everything from tracking and analyzing trunk movement data to creating all the graphs in their journal article, Schulz said. MATLAB was especially beneficial for visualizations, he added. Their team included anatomists and doctors from the Icahn School of Medicine at Mount Sinai, Zoo Atlanta’s head of research, and head veterinarian. MATLAB enabled the mechanical engineers to explain their results to collaborators and to the Zoo Atlanta keepers accurately and cleanly.

Georgia Tech undergraduates contributed to the research through a shared MATLAB space. By reviewing the analysis and commenting on it, the students helped Schulz communicate better with team members: “These are the results we’re getting. Does this make sense to you as somebody reading this for the first time?”

Advancing the Field

The study made news headlines after it was published, appearing in The Atlantic, Smithsonian magazine, and The New York Times and on NPR.

Robots inspired by elephant trunks already exist, including one for autonomous underwater ship refueling, but the Georgia Tech mechanical engineers see opportunities to inspire designs for new, more effective robotics.

Just look at what elephants can achieve with their nostrils and lung capacity. Kelly vacuumed water into her trunk at the mindboggling rate of three liters (0.79 gallons) per second and lifted fragile tortillas without any breakage. Schulz observed that few suction-assisted robotic devices can generate a vacuum both above and below the water.

A) Time course of the water flow rate. Closed points indicate mean values and shaded regions show standard deviation. B) Schematic of an elephant trunk applying suction to a flat object of mass mt. The red arrows denote the direction of flow. Nostrils of radius a generate a pressure of P0, which generates a flow u0 in the nasal cavity, and a flow u outside the trunk, which rapidly decays with distance L. (Image credit: Andrew Schulz)

Insights gleaned from the African elephants at Zoo Atlanta could transform other types of technology. Soft robots, in particular, still have key constraints such as how much they can bend, move, and elongate, Schulz said. “Elephant trunks are purely muscle — no bones, no joints — so they’re really good inspiration for this new realm of soft robots.”

In a World Elephant Week live streaming video presentation, Schulz showed an image of magnified elephant trunk skin, revealing the dense fiber array that makes it almost like flexible Kevlar. “Using this, we can look at creating new robots or even creating new materials,” he told the audience.

But the researchers don’t want to get ahead of themselves. “We need to build the groundwork of how the trunk moves so we can understand its actual limitations and capabilities before we start to investigate material and skin,” Schulz said.

Their experiments could also improve the ways that humans support and protect wild elephants. In March 2021, the International Union for Conservation of Nature put the African forest elephant on its critically endangered list alongside the African savannah elephant, citing decades of ivory poaching and habitat loss.

“Bioinspiration is great, but if the species that we want to get inspiration from are no longer here, then it generates a big challenge. It’s important to work on both research and conservation simultaneously.”- Andrew Schulz, Ph.D. student at Georgia Tech

“It’s well-documented that the more a species is studied, the more conservation focuses on them,” Schulz said. “We’re working on different techniques to make sure that the elephants are going to live into future generations. Right now, they’re on track to become extinct by 2038.”

Stronger elephant trunk anatomy understanding might improve veterinary care for the animals. Previously, many believed that elephants used their trunks like a straw. Now that we know about nostril dilation, it makes sense why an elephant with wounds caused by poaching would need to immerse the trunk in water up to the hole to generate backpressure, Schulz said. Vets could do additional examinations using ultrasound, including whether the animal has the muscular support to generate suction for drinking.

Schulz wants his work to spark broader conversations about elephants, what the animals can do, and how to address myriad threats to their survival in the wild.

“Bioinspiration is great, but if the species that we want to get inspiration from are no longer here, then it generates a big challenge,” Schulz said. “It’s important to work on both research and conservation simultaneously.” The Hu Laboratory will continue to focus on those areas well into the future, he added.

In the meantime, he still goes to see Kelly nearly every week. Schulz anticipates that she and her fellow African elephants at Zoo Atlanta still have much more to teach their human friends.

--

--