“The sea turtle’s unique body shape, the morphology of their flippers and their varied gait patterns makes them very adaptable,” said Yasemin Ozkan-Aydin, assistant professor of electrical engineering at the University of Notre Dame and a roboticist.

“Mimicking this adaptability is challenging because it requires an intricate understanding of how morphology, flexibility and gait interact with the environment. Studying how sea turtles adapt their gaits to traverse complex and varied terrains can help us design more versatile robots.”

Ozkan-Aydin, electrical engineering doctoral student Nnamdi Chikere and undergraduate John Simon McElroy, a Naughton Fellow from University College Dublin, have designed and built a robotic sea turtle, which they are testing in varied environments on Notre Dame’s campus. Their robot mimics a real sea turtle’s propulsion: its front flippers move it forward while its smaller hind flippers allow it to change direction.

The key components of their turtle-robot are an oval-shaped body, four independently radio-controlled flippers, an electronic onboard control unit, a multi-sensor device and a battery. The body frame and flipper connectors are 3D printed using a rigid polymer. The flippers are molded from silicone to provide both flexibility and stiffness.

The robot was designed using data from zoological studies on the morphology, gait patterns and flipper flexibility of multiple sea turtle species. “To maximize adaptability and versatility, we studied the locomotion patterns of different species and incorporated the most effective aspects from each,” Ozkan-Aydin said.

Ozkan-Aydin modeled the robot on the size and structure of sea turtle hatchlings. Sea turtle babies are particularly vulnerable — only one in a thousand survive to adulthood. Hatchlings must run a gauntlet of predator sea birds on their journey from nest to ocean, and that journey has become more perilous by a disorienting landscape of beach development and debris.

“Our hope is to use these baby sea turtle robots to safely guide sea turtle hatchlings to the ocean and minimize the risks they face during this critical period,” Ozkan-Aydin said.

Originally published by Karla Cruise at news.nd.edu on August 07, 2023.

The World

Omar Duwaji

Scientists have been working to develop robotic worms for a few decades. Lately, they've been getting much closer to achieving that goal. The World's host Carolyn Beeler spoke with Yasemin Ozkan-Aydin, an electrical engineer at the University of Notre Dame, about how these robotic worms can be applied and when we might see them in action.

Technology of Business reporter

For decades, scientists have been developing soft robots inspired by a creature often taken for granted - the humble earthworm.

Though specific features will vary by species, many earthworms are excellent burrowers, and can bend with ease.

"They are very flexible and move through spaces that might be difficult to access," comments Elsa Arrázola-Vásquez, who researches soil management at the Swedish University of Agricultural Sciences.

In other words, earthworms can do things that many machines still cannot.

The progress in replicating these functions robotically has been incremental. There have been innovations in mimicking earthworm setae (bristles), which help worms to anchor. Plus, there's been progress copying the fluid in their segments which, among other functions, help them to move.

Capturing the distinctive movement of the earthworm is the latest advance.

In the Soft Robotics group at the Italian Institute of Technology (IIT), researchers have developed a robot that essentially bulges in and out at the sides while it stretches and contracts in length.

This design is novel, according to Riddhi Das, a mechanical engineer at IIT, as it uses positive and negative pressure to generate force which is directed outwards and along the length of the robotic worm.

The innovation means his creation more closely represents how an earthworm's muscles move, and allows for more varied movement.

His earthworm robot is about the length and weight of a light dumbbell. It's filled with gel which allows the researchers to better approximate the earthworm's radial movements, compared to other types of liquid. And though it's not as fast on a flat surface as some previous designs, it's able to move deeper through artificial soil.One person who knows how hard it is to build an earthworm-type robot that can burrow is Yasemin Ozkan-Aydin, an electrical engineer at the University of Notre Dame in the US.

She has worked on four earthworm robot designs, drawing on observations of real earthworms.

She says the IIT group's innovation is "very important" in the world of robotics, as each segment of their earthworm robot has the capability to expand in two directions. That allows it to create an earthworm-type locomotion, which is like a wave of contraction and expansion that moves along its body.

Like the other projects in her lab at IIT, this is genuinely bioinspired, says biologist-turned-roboticist Barbara Mazzolai. This means that developing the prototype required a fundamental understanding of earthworm biology, rather than just mimicking its shape.

And those biological principles had to bring some useful function to the robot. In the case of bioinspired earthworm robots, one useful aspect to replicate is an earthworm's soft yet strong structure.

One key difference remains the size. With a diameter of 4cm and a length of 45 cm, the IIT robot is considerably larger than an actual earthworm. Earthworm robots typically have pumps or other systems for movement that add to their bulk. This limits the potential applications in endoscopy (the use of tubes to examine internal organs).

Still, this robot isn't as big as the rugged tunnelling robot created by the multinational company GE, one of the few earthworm robots being developed for near-term commercial applications.

It's a self-propelling, extremely flexible, highly steerable robot, according to Deepak Trivedi, a mechanical engineer at GE Research.

"If you look at the basic building blocks of these robots, these are pneumatic artificial muscles, which are essentially rubber with a cleverly designed fibre mesh around it," Mr Trivedi explains.

This research line was inspired by a call for tactical tunnelling solutions from the Defense Advanced Research Projects Agency (Darpa) in the US.

"Earthworms can do this tunnelling in a very stealthy way," Mr Trivedi points out. While that Darpa programme has now ended, GE is continuing to work with the US Department of Defense on specific use cases in tunnelling and navigating.

They are also seeking out commercial clients. Their robot, which has a diameter of about 10cm, is unusual in being able to create its own tunnels and having been tested in real soil of different types.

The GE researchers believe that it would be useful for installing underground utility infrastructure, in a less environmentally damaging manner than some conventional drilling. They're aiming for a lower cost as well.

"We see a real commercial opportunity for this," says John Lizzi, who leads the Robotics and Autonomous Systems division at GE Research. He believes key areas include fibre internet, electrical power and charging infrastructure for electric vehicles.

However, GE is limited in what it can publish and publicly disclose about this research, given the military funding. And of course not every roboticist wants to work towards military applications.

Apart from those, eventually earthworm-like robots could also be applied in areas like mining, agricultural sensing, and planetary excavation.

An especially important use could be in search-and-rescue. Prof Ozkan-Aydin talks of the recent earthquakes that devastated Turkey, her country of origin. A tiny wriggling robot with a camera attached could have been useful there for determining where to concentrate rescue efforts, without disturbing the ground.

But plenty of research will need to be done first. Certain essential features of earthworm biology - like the mucus they secrete to lubricate their passage through soil and keep themselves from drying out - are challenging to incorporate into a robot.

"Because it's a natural system and it has evolved for so many years, it's very difficult to replicate," acknowledges Mr Das of IIT.

So nobody is going to mistake one of these robots for a living earthworm. And some people have been overly optimistic about bioinspired systems in the past.

"I'm a bit disappointed that more artificial muscle technology hasn't made it across the 'R&D valley of death'," admits Kellar Autumn, a biologist at Lewis & Clark College in the US.

Like his work on gecko-inspired adhesives, he believes that artificial muscles are following the "10/10 rule of innovation" - 10 years to discover, and another 10 to reach a market.

If these earthworm-inspired robots do eventually reach a market, they could someday be tunnelling away beneath our feet, helping to lay down the equipment that keeps our societies functioning.

YASEMIN OZKAN-AYDIN/UNIVERSITY OF NOTRE DAME

One of the creepiest encounters I’ve ever had came 25 years ago, when I moved into my first house. Turning on the bathroom light early one morning, a gigantic centipede—with its blur of legs—ran up and over my foot without breaking stride. Now, even the sight of one sends shivers up my spine. Not so for Daniel Goldman, a biological physicist at the Georgia Institute of Technology (Georgia Tech) whose lab has just worked out how these invertebrates are so adept at scampering across feet, sand, soil, rocks—and even water. What’s more, he and his colleagues reported last month at the virtual meeting of the Society of Integrative and Comparative Biology that they have created a centipede robot that might one day scutter through farmers’ fields to take out troublesome weeds.

“Their results hammer [home] the point that flexibility [makes animals] capable of a broad range of behaviors,” says Jake Socha, a comparative biomechanist at Virginia Polytechnic Institute and State University (Virginia Tech) who was not involved with the work. The findings, he adds, reveal the previously unknown principles of centipede movement.

Though centipedes don’t typically have the 100 legs they are named for, they often have dozens of pairs, one per body segment. That makes them long as well as leggy—and capable of a variety of motions, says Matthew McHenry, a biomechanist at the University of California (UC), Irvine. The primitive arthropods have “speed, elegance, and efficiency that are the envy of engineers,” McHenry says. Despite that, our intuition about how they move is “often wrong,” Socha says.

Goldman had long been fascinated by centipedes, but analyzing their movements was nearly impossible, because they have too many body segments and legs to track. His interest was piqued when his postdoc, electrical engineer Yasemin Ozkan-Aydin, discovered the value of having numerous body segments and legs. When Ozkan-Aydin, now a robotics scientist at the University of Notre Dame, hooked two or three four-legged robots together, the longer machines could cross wider gaps and clamber over bigger obstacles, even outside on natural terrain, she and colleagues reported last year in Science Robotics. So Goldman’s lab went “all in” on centipede movement.

Another student, Georgia Tech undergraduate Eva Erickson, looked at how centipedes change the way they run. Many animals, including horses and people, alter their gait—the way their legs move relative to their body—as they increase their speed. But, using a sophisticated video tracking program called DeepLabCut, Erikson found the centipede Scolopocryptops sexspinosus instead changes its gait to match the challenges of its terrain.

Normally, S. sexspinousus’s legs move in a wave—like fingers drumming on a table. But sometimes, the direction of the wave changes. On flat surfaces, the wave starts with the last leg and travels headward. But when the going gets tough, the wave reverses, with the front leg moving first to establish a foothold, Erickson reported at the meeting. After that, each leg follows in kind, landing in the exact same spot as the previous leg. The taller the obstacles, the more likely the centipede is to adopt this so-called retrograde locomotion, Erickson said.

Should they wind up in the water, centipedes can also “swim” to rescue themselves. When Kelimar Diaz, a graduate student at Georgia Tech, investigated the behavior, she, too, found centipedes adjusted their movements to their environment. Videos reveal the species Lithobious forficatus starts to “swim” by flailing its legs. But then it quickly twists its body from side to side, at which point it moves forward (see video, below). “Somehow, the centipedes know that undulating their bodies generates the right forces,” Diaz reported at the meeting.

Another graduate student in the lab, Baxi Chong, has analyzed how separate waves in the legs and the bodies sync up. To predict which combination of body waves and stepping works best, he used a mathematical modeling technique first proposed by particle physicists. The model produced numerous combinations of leg and body waves, which he then tested in a centipede robot built solely for understanding how these animals move.

First, Chong programmed the robot to move its legs and body in sync. But he soon found the robot was faster when there was a lag between the two waves, he reported at the meeting. Further testing revealed some combinations made the robot move backward. These experiments, says UC Berkeley integrative biologist Robert Full, have “unraveled the secrets” of centipede coordination.

Subsequent work by Goldman’s group has found the robot works better if its legs are jointed and the body segments are pliable. “Being soft and squishy has lots of advantages both for animals and robots,” Socha says. “And animals that are long and skinny can take advantage of combinations of waves” to move, he adds.

Goldman’s latest robots are flexible, and they combine the fleet-footedness of a cockroach with the reach of a snake. The next step, he says, is to train them to do practical tasks, such as weed identification and eradication. Working with Philip Benfey, a plant development biologist at Duke University, Goldman plans to outfit his robots with artificial intelligence software that can single out individual weeds; the duo envisions robots that could then inject herbicide or use a laser or electrical discharge to kill them.

And if Ozkan-Aydin has her way, one day swarms of robotic centipedes will help humans with other tasks: planting and harvesting crops, moving goods, monitoring the environment, and even space exploration. That’s fine with me, as long as they stay off my feet.

In the wild, a worm blob looks like any other mud ball lolling around the bottom of a pond. But if you poke an unassuming worm blob, it will respond in a way a mud ball never would, wriggling out into a noodly shape that a Pastafarian might mistake for something divine.

This is how Saad Bhamla discovered his first worm blob, in a pond in California. “As you poke it with a stick, it comes alive,” said Dr. Bhamla, a bioengineer at the Georgia Institute of Technology’s school of chemical and biomolecular engineering. Dr. Bhamla’s encounter with the worm blob haunted him for years (in a good way, he says) until he started his own lab and needed a first project.

California blackworms, soft and slender ropes as surreally red as grocery store meat, often live in seasonal pools. When times are good, a worm is simply a worm, wiggling about on its own. When times are bad, a worm must become a blob, entangling with hundreds or thousands of other worms into a slimy, writhing ball. And, like an animated ball of yarn, the worm blob can move as one unit, meandering away from predators or stress.

“They remain braided and twisted into this cohesive unit that’s crawling around,” said Chantal Nguyen, a postdoctoral associate and physicist at the BioFrontiers Institute at the University of Colorado Boulder.

But how does a worm attain and maintain blobdom? In a recent study in the journal Frontiers in Physics, a group of researchers including Dr. Nguyen and Dr. Bhamla unraveled the secrets of the blob’s ability to move. They did so by creating a computer model of entangled California blackworms.

“It was pretty horrific and pretty shocking, but also kind of beautiful,” said Albert Kao, a postdoctoral fellow studying collective behavior at the Santa Fe Institute in New Mexico, of the worm blobs. The simulation, he added, “lays a path forward for new kinds of models for similarly entangled systems.”

Since time immemorial, people have witnessed groups of animals moving collectively and in unison: starlings flock, fish school, midges swarm, and heavy metal heads mosh. But few people have had the privilege of, or the interest in, observing worm blobs.

A worm blob behaves as a solid and a fluid, like a ball of dough or a glob of shampoo. It only takes around 10 worms to form a coherent blob. A blob of about 100,000 worms resembles a lump of (red) pizza dough. There is no known limit to how many worms can form a blob, except, perhaps, your imagination.

Dr. Ding, who studies blobbing in the much-studied nematode Caenorhabditis elegans, described her C. elegans blobs as “strongly overlapping, like a bowl of spaghetti noodles.” Blackworm blobs “are more like spaghetti noodles dropped on the floor,” she said, frowning, in a Zoom call. “C. elegans is named for being elegant. These ones are just … not.”

But it was precisely this messy splat of blackworm blobs that captured Dr. Bhamla’s heart. To him, the blobs feel like pizza dough flowing through fingers. “But it’s composed of worms,” he said. “Like a nightmare come alive.”

In February, Dr. Bhamla and a group of researchers described the dynamics of worm blobs in the journal Proceedings of the National Academy of Sciences.

But, what about more complex terrain? The capability that would allow land-based search-and-rescue robot swarms to navigate buildings and other disaster areas does not yet exist. Researchers at the Georgia Institute of Technology are working to develop simple, low-cost, legged robots capable of linking and unlinking to accomplish tasks, such as gap traversal, stair climbing, and object transport over uneven terrains.

Working with Daniel Goldman, Dunn Family professor in the School of Physics at Georgia Tech, Yasemin Ozkan-Aydin, a former postdoc in Goldman’s lab and now an assistant professor at the University of Notre Dame, developed “quadruped” robots using easily acquired off-the-shelf technology. Each unit has a 3D-printed, two-segmented chassis and body, four flexible legs, a “passive tail” appendage for additional balance and directional control, touch and light sensors, and a central-body-mounted microprocessor. A magnetic connector allows for docking and cooperative behaviors. The research team recently published its work in Science Robotics.

Reconfigurable swarm robots have been used in prior research into terrain solutions. But these units have limited motive abilities, require human intervention, and lack the desired ease of scale-up for manufacturing and use in a timely and cost-effective manner.  

The new research takes advantage of enhanced mobility in multi-legged robots, mechanical intelligence (passive flexible legs and tail), and simple cooperative effect.  Individualrobots perform simple or small tasks, but if the task is beyond the capability of the one unit, a team of robots physically connect to each other and form a larger multi-legged system to collectively overcome issues.

Additionally, the team’s use of widely available technologies – like 3D printers – could lead to cost-effective and rapid development of terrestrial robotic teams that can collaboratively move heavy or dangerous objects. They could also be used for search-and-rescue operations, environmental monitoring, or even space exploration. 

Ozkan-Aydin sees the work as a starting point in the development of land-based robot swarms with the ability to climb hills, overcome obstacles, and move on rough terrain by maintaining stability in a truly autonomous fashion.

“The development of physical connection between individual robot units can improve the mobility of an entire terrestrial collective system, and help prevent failures when attempting a task,” said Ozkan-Aydin. “Furthermore, the minimalist and modular robotic approach taken in this study can provide a low-cost platform for testing or generating new hypotheses for biological research.”

“Self-reconfigurable multilegged robot swarms collectively accomplish challenging terradynamic tasks”, Ozkan-Aydin and Daniel I. Goldman, Science Robotics; 28 Jul 2021: Vol. 6, Issue 56, eabf1628, DOI: 10.1126/scirobotics.abf1628

In research published in Science Robotics, Ozkan-Aydin presents how she was able to build multi-legged robots capable of maneuvering in challenging environments and accomplishing difficult tasks collectively, mimicking their natural-world counterparts.

“Legged robots can navigate challenging environments such as rough terrain and tight spaces, and the use of limbs offers effective body support, enables rapid maneuverability and facilitates obstacle crossing,” Ozkan-Aydin said. “However, legged robots face unique mobility challenges in terrestrial environments, which results in reduced locomotor performance.”

For the study, Ozkan-Aydin said, she hypothesized that a physical connection between individual robots could enhance the mobility of a terrestrial legged collective system. Individual robots performed simple or small tasks such as moving over a smooth surface or carrying a light object, but if the task was beyond the capability of the single unit, the robots physically connected to each other to form a larger multi-legged system and collectively overcome issues.

“When ants collect or transport objects, if one comes upon an obstacle, the group works collectively to overcome that obstacle. If there’s a gap in the path, for example, they will form a bridge so the other ants can travel across — and that is the inspiration for this study,” she said. “Through robotics we’re able to gain a better understanding of the dynamics and collective behaviors of these biological systems and explore how we might be able to use this kind of technology in the future.”

Using a 3D printer, Ozkan-Aydin built four-legged robots measuring 15 to 20 centimeters, or roughly 6 to 8 inches, in length. Each was equipped with a lithium polymer battery, microcontroller and three sensors — a light sensor at the front and two magnetic touch sensors at the front and back, allowing the robots to connect to one another. Four flexible legs reduced the need for additional sensors and parts and gave the robots a level of mechanical intelligence, which helped when interacting with rough or uneven terrain.

“You don’t need additional sensors to detect obstacles because the flexibility in the legs helps the robot to move right past them,” said Ozkan-Aydin. “They can test for gaps in a path, building a bridge with their bodies; move objects individually; or connect to move objects collectively in different types of environments, not dissimilar to ants.”

Ozkan-Aydin began her research for the study in early 2020, when much of the country was shut down due to the COVID-19 pandemic. After printing each robot, she built each one and conducted her experiments at home, in her yard or at the playground with her son. The robots were tested over grass, mulch, leaves and acorns. Flat-ground experiments were conducted over particle board, and she built stairs using insulation foam. The robots were also tested over shag carpeting, and rectangular wooden blocks were glued to particle board to serve as rough terrain.

When an individual unit became stuck, a signal was sent to additional robots, which linked together to provide support to successfully traverse obstacles while working collectively.

Ozkan-Aydin says there are still improvements to be made on her design. But she expects the study’s findings will inform the design of low-cost legged swarms that can adapt to unforeseen situations and perform real-world cooperative tasks such as search-and-rescue operations, collective object transport, space exploration and environmental monitoring. Her research will focus on improving the control, sensing and power capabilities of the system, which are essential for real-world locomotion and problem-solving — and she plans to use this system to explore the collective dynamics of insects such as ants and termites.

“For functional swarm systems, the battery technology needs to be improved,” she said. “We need small batteries that can provide more power, ideally lasting more than 10 hours. Otherwise, using this type of system in the real world isn’t sustainable.” Additional limitations include the need for more sensors and more powerful motors — while keeping the size of the robots small.

“You need to think about how the robots would function in the real world, so you need to think about how much power is required, the size of the battery you use. Everything is limited so you need to make decisions with every part of the machine.”

Daniel I. Goldman at the Georgia Institute of Technology co-authored the study.

Contact: Jessica Sieff, assistant director of media relations, 574-631-3933, jsieff@nd.edu

Originally published by Jessica Sieff at news.nd.edu on October 18, 2021.