A team of researchers has beaten its own record for the fastest swimming soft robot, drawing inspiration from manta rays to improve their ability to control the robot’s movement in the water.

“Two years ago, we demonstrated an aquatic soft robot that was able to reach average speeds of 3.74 body lengths per second,” says Jie Yin, corresponding author of a paper on the work and an associate professor of mechanical and aerospace engineering at North Carolina State University.

“This is a highly engineered design, but the fundamental concepts are fairly simple.”

“We have improved on that design. Our new soft robot is more energy efficient and reaches a speed of 6.8 body lengths per second. In addition, the previous model could only swim on the surface of the water. Our new robot is capable of swimming up and down throughout the water column.”

The soft robot has fins shaped like those of a manta ray, and is made of a material that is stable when the fins are spread wide. The fins are attached to a flexible, silicone body that contains a chamber that can be pumped full of air. Inflating the air chamber forces the fins to bend—similar to the down stroke when a manta flaps its fins. When the air is let out of the chamber, the fins spontaneously snap back into their initial position.

“Pumping air into the chamber introduces energy into the system,” says Haitao Qing, first author of the paper and a PhD student at NC State. “The fins want to return to their stable state, so releasing the air also releases the energy in the fins. That means we only need one actuator for the robot and allows for more rapid actuation.”

Studying the fluid dynamics of manta rays also played a key role in controlling the vertical movement of the soft robot.

“We observed the swimming motion of manta rays and were able to mimic that behavior in order to control whether the robot swims toward the surface, swims downward, or maintains its position in the water column,” says Jiacheng Guo, coauthor of the paper and a PhD student at the University of Virginia.

“When manta rays swim, they produce two jets of water that move them forward. Mantas alter their trajectory by altering their swimming motion. We adopted a similar technique for controlling the vertical movement of this swimming robot. We’re still working on techniques that will give us fine control over lateral movements.”

“Specifically, simulations and experiments showed us that the downward jet produced by our robot is more powerful than its upward jet,” says Yuanhang Zhu, coauthor of the paper and an assistant professor of mechanical engineering at the University of California, Riverside.

“If the robot flaps its fins quickly, it will rise upward. But if we slow down the actuation frequency, this allows the robot to sink slightly in between flapping its fins—allowing it to either dive downward or swim at the same depth.”

“Another factor that comes into play is that we are powering this robot with compressed air,” Qing says. “That’s relevant because when the robot’s fins are at rest, the air chamber is empty, reducing the robot’s buoyancy. And when the robot is flapping its fins slowly, the fins are at rest more often. In other words, the faster the robot flaps its fins, the more time the air chamber is full, making it more buoyant.”

The researchers have demonstrated the soft robot’s functionality in two different ways. First, one iteration of the robot was able to navigate a course of obstacles arrayed on the surface and floor of a water tank. Second, the researchers demonstrated that the untethered robot was capable of hauling a payload on the surface of the water, including its own air and power source.

“This is a highly engineered design, but the fundamental concepts are fairly simple,” Yin says. “And with only a single actuation input, our robot can navigate a complex vertical environment. We are now working on improving lateral movement, and exploring other modes of actuation, which will significantly enhance this system’s capabilities. Our goal is to do this with a design that retains that elegant simplicity.”

The paper appears in the journal Science Advances.

Additional coauthors are from NC State and the University of Virginia.

Support for this work came from the National Science Foundation and the Office of Naval Research.

Source: North Carolina State University

Related

Watch: Robot snake goes where people can’t

Robot cuts off its own limb to keep moving

Jelly sea creature ‘jet propulsion’ could give robots a boost

A new robot can sever its own limbs to escape tricky situations.

Self-amputation may seem like a drastic move, but it’s a survival tactic that’s proved particularly handy for numerous creatures.

Roboticists have drawn inspiration from lizards, crabs, and other animals who shed parts of themselves without looking back, all for the purpose of moving forward.

Outside the lab, a robot could face many potential perils: A falling tree limb, perhaps, or getting stuck under a rock on a search-and-rescue mission. In most cases, that would be it for the robot.

But the Yale University lab of Rebecca Kramer-Bottiglio has developed a technology that allows a robot to selectively disconnect its limbs and free itself from such potential pitfalls.

Conversely, it allows separate robots to join together to take on tasks that they couldn’t do on their own.

Critical to the technology is a material that the roboticists invented in the lab that they call a bicontinuous thermoplastic elastomer. The thermoplastic they use is a rubbery solid at room temperature that melts into a liquid at about 284 °F. It’s infused in a foam-like structure made from silicone that holds the thermoplastic in place when it’s melted into a liquid.

How it works: Two silicone bodies each have a layer of bicontinuous thermoplastic elastomer on their exposed surfaces. The foams are heated so that the thermoplastic melts into liquid. The silicone matrix holds the molten material like a sponge, preventing it from flowing out. Upon contact of the two parts, the molten material combines into a continuous liquid mass. The material then cools and solidifies to connect the two parts. And to disconnect, the joint is heated so that the material melts and weakens, allowing the two parts to separate easily.

“So if the robot is doing its normal operations and walking around the wild, but then something happens to one of its legs—a big rock falls on it, for example—normally the whole robot would be stuck if it were cast in whole,” says Bilige Yang, a PhD student and the lead author of the work.

“But because we have the ability to melt away and weaken this joint where the material is, the rest of the robot will be able to walk away without its amputated leg.”

It’s a similar tactic used by lizards; if an attacker grabs its tail, the lizard automatically releases the limb and scurries to safety. And a crab will shed an injured appendage that’s slowing it down. But the Kramer-Bottiglio lab also points to the world of ants as a source of inspiration. For instance, many ants can link up to form a bridge to cross a gap on the forest floor or form into the shape of a ball to float on water.

Yang points to two small robotic devices on a counter in the lab, and a gap too wide for either of them to safely cross alone.

“If each individual robot tried to cross the gap, it would just fall through,” he says. “But if you have a few of them together, they can make it across. You can imagine this in different types of search-and-rescue missions where the robot will be able to navigate these types of scenarios much better.”

Up next, the research team will apply this technology to many other soft robots they’ve developed in the laboratory.

“Our material not only aids in robot survival—it enables dynamic shape-change,” Kramer-Bottiglio says. “Robotic modules can self-reconfigure into different morphologies to perform tasks that demand specific shapes and behaviors.”

A paper on the robot appears in Advanced Materials.

Source: Yale

Related

Origami-inspired ‘transformer bot’ morphs into 1,000 shapes

Stretchy ‘skin’ could give robots sensitivity of human touch

Better fake muscles give robot fish real kick

Researchers have developed artificial muscles for robot motion.

Their solution offers several advantages over previous technologies: it can be used wherever robots need to be soft rather than rigid or where they need more sensitivity when interacting with their environment.

Many roboticists dream of building robots that are not just a combination of metal or other hard materials and motors but also softer and more adaptable. Soft robots could interact with their environment in a completely different way; for example, they could cushion impacts the way human limbs do, or grasp an object delicately.

This would also offer benefits regarding energy consumption: robot motion today usually requires a lot of energy to maintain a position, whereas soft systems could store energy well, too. So, what could be more obvious than to take the human muscle as a model and attempt to recreate it?

Biological inspiration

The functioning of artificial muscles is based on biology. Like their natural counterparts, artificial muscles contract in response to an electrical impulse. However, the artificial muscles consist not of cells and fibers but of a pouch filled with a liquid (usually oil), the shell of which is partially covered in electrodes.

When these electrodes receive an electrical voltage, they draw together and push the liquid into the rest of the pouch, which flexes and is thus capable of lifting a weight. A single pouch is analogous to a short bundle of muscle fibers; several of these can be connected to form a complete propulsion element, which is also referred to as an actuator or simply as an artificial muscle.

The idea of developing artificial muscles is not new, but until now, there has been a major obstacle to realizing it: electrostatic actuators worked only with extremely high voltages of around 6,000 to 10,000 volts. This requirement had several ramifications: for instance, the muscles had to be connected to large, heavy voltage amplifiers; they did not work in water; and they weren’t entirely safe for humans.

Robert Katzschmann, a robotics professor at ETH Zurich, together with Stephan-Daniel Gravert, Elia Varini, and other colleagues have now developed a new solution.

The HALVE of it

Gravert, who works as a scientific assistant in Katzschmann’s lab, has designed a shell for the pouch. The researchers call the new artificial muscles HALVE actuators, where HALVE stands for “hydraulically amplified low-voltage electrostatic”.

“In other actuators, the electrodes are on the outside of the shell. In ours, the shell consists of different layers,” says Gravert.

“We took a high-permittivity ferroelectric material, i.e., one that can store relatively large amounts of electrical energy, and combined it with a layer of electrodes. Next, we coated it with a polymer shell that has excellent mechanical properties and makes the pouch more stable,” Gravert explains.

This meant the researchers could reduce the required voltage, because the much higher permittivity of the ferroelectric material allows large forces despite low voltage. Not only did Gravert and Varini develop the shell for the HALVE actuators together, but they also built the actuators themselves in the lab to use in two robots.

One of these robotic examples is an 11-centimeter-tall gripper (about 4.3 inches tall) with two fingers. Each finger is moved by three series-connected pouches of the HALVE actuator. A small battery-operated power supply provides the robot with 900 volts. Together, the battery and power supply weigh just 15 grams (about 0.5 ounces). The entire gripper, including the power and control electronics, weighs 45 grams (about 1.58 ounces). The gripper can grip a smooth plastic object firmly enough to support its own weight when the object is lifted into the air with a cord.

“This example excellently demonstrates how small, light and efficient the HALVE actuators are. It also means that we’ve taken a huge step closer to our goal of creating integrated muscle-operated systems,” Katzschmann says with satisfaction.

Diving into the future

The second object is a fish-like swimmer, almost 30 centimeters long (about 11.8 inches), that can move smoothly through the water. It consists of a “head” containing the electronics and a flexible “body” to which the HALVE actuators are attached. These actuators move alternately in a rhythm that produces the swimming motion. The autonomous fish can go from a standstill to a speed of three centimeters per second in 14 seconds—and that’s in normal tap water.

This second example is important because it demonstrates another new feature of the HALVE actuators: as the electrodes no longer sit unprotected outside the shell, the artificial muscles are now waterproof and can also be used in conductive liquids.

“The fish illustrates a general advantage of these actuators—the electrodes are protected from the environment and, conversely, the environment is protected from the electrodes. So, you can operate these electrostatic actuators in water or touch them, for example,” Katzschmann explains.

And the layered structure of the pouches has another advantage: the new actuators are much more robust than other artificial muscles.

Ideally, the pouches should be able to achieve a great deal of motion and do it quickly. However, even the smallest production error, such as a speck of dust between the electrodes, can lead to an electrical breakdown—a kind of mini lightning strike.

“When this happened in earlier models, the electrode would burn, creating a hole in the shell. This allowed the liquid to escape and rendered the actuator useless,” Gravert says. This problem is solved in the HALVE actuators because a single hole essentially closes itself due to the protective plastic outer layer. As a result, the pouch usually remains fully functional even after an electrical breakdown.

The two researchers are clearly delighted to have taken the development of artificial muscles a decisive step forward, but they are also realistic.

As Katzschmann says, “Now we have to ready this technology for larger-scale production, and we can’t do that here in the ETH lab. Without giving too much away, I can say that we’re already registering interest from companies that would like to work with us.”

For example, artificial muscles could one day be used in novel robots, prostheses, or wearables; in other words, in technologies that are worn on the human body.

The research appears in Science Advances.

Source: ETH Zurich

Related

Bug robots might be the smallest and fastest ever

Twisted ring robot spins like a record to get around

Team prints robotic hand with bones, ligaments, tendons