Researchers at the University of Virginia have developed a new tuna-inspired robot that can flap its tail as quickly as actual tuna (seven times per second) and swim at speeds of up to 1.5 mph, or two body lengths per second.
The team behind this invention explains how they created this tuna-like robot, which can tighten or relax its tail joint to move at different speeds, in a recent study published in Science Robotics.
The tuna bot was able to attain high speeds while conserving energy thanks to this technology, which might one day be used to improve swimming robots and even underwater vehicles.
Many roboticists have been experimenting with their robot fish.
“We are unique in that we investigate the physics [of genuine fish] rather than merely imitating the design or schematic to construct something that looks like a real robot fish,” says Qiang Zhong, a UVA postdoctoral researcher, and the paper’s first author.
The researchers utilized flow physics to infer the flow physics of real tuna motions, which they then used to guide their design.
Fish use their muscles not just to create motion, but also to control the flexibility and stiffness of their tails to swim effectively.
They have the ability to adjust their rigidity to be as floppy as a piece of paper or as solid as a branch.
This is comparable to how automobiles increase their efficiency by shifting gears at various speeds.
Tuna have tendons that span the length of their bodies, and they may stiffen their tails to swim faster by exerting stress on these tendons.
With springs and motors, the researchers attempted to replicate this muscle-tendon system.
Zhong explains, “The study we do is to try to grasp the proportion theory of fish.”
“We aren’t simply trying to make something that looks like a fish; we want to take the inspiration from the fish, figure out what’s going on, and apply this approach to whatever we choose.”
The result was a simpler robofish with high-frequency actuators (which approximated motion) and a spring mechanism that translated spring force to tail joint stiffness.
Nature as a source of inspiration for human-centered machines
The tuna robot is one of a small group of biomimicry, or biomimetic robots, which copy animals.
Robots that were influenced by octopuses, manta rays, cheetahs, pigeons, and bats, to mention a few, were among the previous submissions.
Researchers are increasingly turning to nature for inspiration on how to create better robots.
In fact, in 2019, another UVA team exhibited their tuna robot at Science Robotics.
Soft robots may also adjust their stiffness using systems such as springs, actuators, artificial tendons, and motors.
Soft robots with changeable flexibility are the best capable of smoothly changing speeds.
“They’ve attempted to come up with stiffness adjusting systems in other robots, such as walking robots, so there’s precedence.
“We had to figure out how to get it into a fish,” says Daniel Quinn, an assistant professor of mechanical and aerospace engineering at UVA and one of the paper’s authors.
Quinn claims that the fascination in fish bots derives from the fact that fish are so nimble compared to the greatest underwater vehicles now available, which are clunky and only capable of traveling at one cruise speed in a straight line.
He explains, “They have control propellers to position themselves, but they’re not intended to do loop-de-loops as fish do.”
“If you need to leave from a coastal outpost, swim for miles to an oil spill quickly, then slow down and make these tight turns to take measurements or stand very still—right now, our vehicles are terrible at that kind of multi-modal mission.”
Additionally, when operating in shallow waters, seaweeds and other marine gunk can easily jam propellers.
Not only would a fish-like vehicle be able to glide over the underwater environment, but it could also run at a lower frequency than propellers, resulting in far less noise.
“It’s a lot more ecologically friendly,” Zhong adds, “especially if you want to perform some ocean resource research and swim around coral or other fish.”
The study might lead to the next generation of underwater vehicles, thanks to funding from the Office of Naval Research and the National Science Foundation.
How is tuna prepared?
The tuna robot is divided into two parts: a drag-generating “head” and a thrust-generating “tail.” An actuator sits in the front of the tail and drives it by creating a back-and-forth motion that reverberates throughout the joint.
Quinn adds that the robot tuna is suspended from a rig and set on a platform of air bushings, allowing it to “slide about like it’s on an air hockey table.”
“That means we can examine how this fish moves about independently while also taking extremely accurate data since it isn’t fully untethered.”
The fish model’s head, tail frame, tail shell, tail joint, tail fin, and tail fin connector were all 3D printed before being assembled.
Except for the tail joint, every item was 3D-printed in waterproof nylon.
The space between the tail frame and the shell was filled with silicone gel.
To withstand tension, the tail joint and tail fin connector were 3D-printed in stainless steel.
An automated, pre-programmed code controls the tuna robot.
The tuna was tested in a water channel that acts like a treadmill for the swimming robot, which allowed it to adjust its swimming style in real-time based on environmental conditions in a closed loop.
Quinn adds, “We just ratchet up the pace so it has to swim faster and faster as the tunnel replicates faster and faster flow speeds.”
The tuna robot was placed on “missions” such as a serpentine, 650-foot dash in which it gradually increased its pace from slow to rapid.
The tuna robot, while amazing in the end, is merely a prototype of the technology that the team intends to develop.
“As you can see, our tail remains stiff. The tail of a genuine fish, on the other hand, is comprised of muscle,” Zhong explains.
“They have a near-infinite range of options.”
One of the main reasons why fish are so adaptable is because of this. “Right now, our system is constrained.
We can’t do complex movements like sea stars or brittle fish.
To accomplish so, he says, you must increase and improve the degrees of freedom.
“In the future, we aim to regulate the entire tail stiffness with smart materials like artificial muscles.
I could expand this to the entire fish if the artificial muscles are powerful enough.
This will make our fish even more realistic, while also reducing the noise created by the other mechanical components.
As a result, our system will be even closer to a natural system.
