Skip to main content

The Evolution and Potential of Soft Robotics

Image Source: Sergey/Stock.adobe.com

By Brandon Lewis for Mouser Electronics

Published June 18, 2024

When most people think of a robot, they picture hard metal and rigid movements. But there’s another kind of robot that defies this popular conception: a robot that is flexible and pliant, moving more like organic life than a machine.

Such robots are referred to as soft robots. Rather than steel and graphite, soft robots are built from materials such as silicone, rubber, and gel, affording them a range and style of motion that is nearly impossible for a traditional machine. Because they move and behave uniquely, soft robots enjoy several advantages over their steel-clad kin, including improved collision resistance and an enhanced capacity for complex motion.

This article briefly examines the history of this exciting robotics subfield, its advantages and disadvantages, and the many compelling applications being explored across multiple sectors.

A Brief History of Soft Robots

Surprisingly, the idea of making robots with more pliable materials has existed for nearly as long as the modern idea of robots. Real-world applications kicked off with the McKibben artificial muscle, developed in 1950.[1] Also known as McKibben air muscles, these devices consist of a flexible pneumatic tube shielded by braided mesh.[2] Although originally developed for orthotics, the pneumatic muscle has been employed in multiple robot designs since its development. McKibben also inspired several technologies that laid the groundwork for the modern field of soft robotics.

The subsequent major development in soft robotics would not occur until the late twentieth century. In 1990, S. Shimachi and M. Matumoto published A Study on Contact Forces of Soft Fingers, an analysis of finger surface deformation, friction, and manipulation.[3]  Just one year later, we saw the development of the first flexible silicone rubber micro-actuator.[4]

In the years that followed, a multitude of new technologies and innovations emerged, including manipulators modeled after tentacles and elephant trunks, artificial muscles comprised of electrostrictive polymer, and fluidic actuators.

Since then, the development of bio-inspired robots and materials has continued to evolve. In 2016, for instance, Harvard University announced the Octobot, the first truly autonomous soft robot.[5] More recently, Cornell University developed a soft robot capable of detecting and healing damage.[6]

Anatomy of a Soft Robot

Soft robots are typically comprised of flexible matter such as fluids, elastomers, or gels through a process known as compliance matching. This process involves tailoring the mechanical properties of the materials to closely match those of the surrounding environment, ensuring even load distribution and minimizing stress. Where actuation is concerned, soft robots may utilize one or more of the following systems:[7]

  • Pneumatic actuators rely on compressed air.
  • Hydraulic actuators rely on fluids such as oil or water.
  • Thermally responsive actuators, also known as shape-memory alloys (SMAs), change shape when exposed to heat.
  • Electroactive polymer actuators use a combination of electrodes, insulating polymers, and conducting polymers to achieve motion.
  • Magnetic actuators generate motion through the application of a magnetic field.
  • Photoresponsive actuators react to light within the visible spectrum by changing shape.[8]
  • Explosive actuators, as the name suggests, use contained explosions for power and locomotion.[9]

The most common process for fabricating and manufacturing soft robots—soft lithographic molding—consists of the following steps:[10]

  1. Manufacturing the internal components, known as the constraint layer. These components provide the necessary rigidity for movement. Traditionally, this may be done through 3D printing or casting.
  2. Modeling the flexible outer layer of the robot after the desired form factor. This is typically done by casting and curing silicon rubber or similar material in a purpose-built mold.
  3. Joining the different layers together via uncured elastomer. After re-curing, the finished actuator is removed from the mold.

More complex soft robots may require additional development or manufacturing steps, including developing the robot's onboard software and firmware, but the core process generally remains the same.

Benefits and Drawbacks of Soft Robotics

Although soft robots have many advantages over complex robots, they are not without weaknesses—nor are they suitable for every application.

Soft robots have enhanced capacity for complex motion and a flexible, adaptable shape that is well-suited for multifunctional applications, particularly those requiring intricate movements or delicate handling. While rigid robots are less flexible and less capable of complex motion, they are more powerful and precise, with a higher load capacity. Due to these characteristics, rigid robots outperform soft robots in heavy industrial applications, particularly those requiring precise movements.[11]

Because soft robots are made from impact-absorbing materials, they're also less prone to collision damage than rigid robots, though they are likelier to take damage from punctures, tears, or cuts. Soft robot materials also tend to be more cost-effective and environmentally friendly than those in rigid robots. Compliance matching for a soft robot is also more accessible, as their materials are more similar to biological matter.

However, two key drawbacks offset these benefits:

  • Soft robots may require specially adapted electronic components that are difficult to source.
  • Soft robots also tend to have higher control and power requirements compared to rigid robots.

One of the most significant advantages soft robots have over rigid robots is that they are better suited to human interaction. While humans may find the motions and appearance of a rigid robot off-putting, they tend to be put at ease by a soft robot's biocompatible design. A 2023 Washington State University study even found that exposure to soft robotics decreased anxiety about working with or being replaced by robots.[12]

Soft Robotics in Healthcare

There is incredible variety in the potential use cases for soft robotics, and new applications are constantly being explored. Given that the McKibben Artificial Muscle was initially developed for medical purposes, it seems fitting that soft robots are particularly useful in healthcare.

Soft robotics' flexibility and pliability could easily be applied to develop complex limb replacements. Thanks to artificial skin developed by Stanford engineers in 2015, these next-generation prosthetics could potentially even mimic an organic limb’s capacity to sense touch.[13] Given that soft robots are already capable of self-repair and autonomous growth, we could eventually even see the development of prosthetics that heal and grow alongside their wearers.[14], [15]

Soft robot designs could also be applied to support suits for patients with limited motion and as wearable physical rehabilitation technology. The wearable robotic exoskeleton is not a new concept, as companies such as ReWalk, Ekso Bionics, and Cybderyne each offer their own take on the technology.[16] Compared to traditional rigid exosuits, soft robotic exosuits have the potential to offer greater comfort and portability while also supporting a more natural range of motion.[17]

One could even take things a step further and use soft robots to replace or reconstruct damaged or missing internal body parts such as organs and muscles. Artificial organs could also be used for medical training, allowing surgeons and other clinicians to practice on more realistic simulations.[18]

Prosthetics and rehabilitation aside, physicians could combine soft robotics and artificial intelligence to implement incredibly sophisticated and minimally invasive diagnostic procedures and surgeries. Professor Sheila Russo, founder and director of the Boston University Material Robotics Laboratory, has made great strides in that regard. Working with students from Harvard Medical School, Russo and her collaborators developed a soft sensor to detect bleeding during colonoscopies alongside a haptic feedback glove to enable easier navigation during endoscopic procedures.[19], [20] More recently, Russo published a paper assessing the use of small-scale soft robots during lung biopsies.[21]

Soft robots could also be used for medication delivery, with gelatin-based devices delivering medicine to specific areas of the body.[22] The potential of this technology goes well beyond medication delivery. More sophisticated soft medical robots could treat disease; in 2023, for example, researchers at the University of Leeds even developed a robot capable of entering the lungs for early lung cancer detection and intervention.[23] Lastly, care providers could deploy humanoid soft robots as assistants for mobility-restricted patients and elders.

Other Potential Soft Robotics Applications

Soft robotic technology is equally promising outside the healthcare space. For first responders and industrial workers, soft exoskeletons such as the exosuit from Harvard's Wyss Institute could provide additional strength and protection.[24] Soft robots may also prove invaluable for search-and-rescue scenarios because they can reach and explore spaces far too small for humans. An added benefit of using soft robots in this way is that it also reduces the risk of injury for first responders—they can send in robots to scout dangerous areas for survivors rather than having to perform the search themselves.

Soft robots capable of self-replication and self-repair, meanwhile, could be used for scientific research into evolutionary biology, ecosystem restoration, construction, and crop management. A plantlike model of a robot, developed by biologist and engineer Barbara Mazzolai, even has the potential to terraform Mars.[25] Mazzolai's “roboplant” behaves similarly to the roots of an organic plant, growing through the soil as it seeks out materials and chemicals.

Mazzolai is also working on another robot model, dubbed the Growbot, inspired by how plants grow aboveground. She envisions a future where robotic plants are used to cultivate soil and grow buildings—perhaps even entire cities—using the resources available on a planet’s surface. Such robots could be sent ahead of astronauts and colonists to prime a landing site for their arrival, creating fully functional structures with working water, electricity, and communications infrastructure.

From Science Fiction to Science Fact

This article has reviewed artificial limbs that are nearly indistinguishable from organic limbs (even to the wearer), robots that are capable of self-repair, and the concept of terraforming other planets through robotic plants. These all seem like concepts taken directly from the pages of a science fiction novel, and perhaps sixty or seventy years ago, that’s all they were.

But times have changed, and prior science fiction has become a reality. These innovations exist within the soft robotics field, innovations on which scientists are already iterating. The developments we are likely to see in the immediate future involve making soft robots more intelligent, pliable, and indistinguishable from biological life while also making them compatible with circular economies.

Potential measures include removing electronic components, more efficient energy storage and generation, and developing robots that imitate a biological organism's lifecycle.[26]

When most people think of machines, they think of rugged steel and rigid components. They don't consider organic bodies or plants. Yet these are perfect machines in their own right—it's only natural that soft robots should imitate them.

It's only natural that as soft robotics continues to evolve, soft robots become increasingly lifelike and indistinguishable from living things.

 

Sources

[1]https://iopscience.iop.org/article/10.1088/1748-3190/acbb48/pdf
[2]https://softroboticstoolkit.com/book/pneumatic-artificial-muscles
[3]https://www.semanticscholar.org/paper/A-study-on-contact-forces-of-soft-fingers.-Shimachi-Matumoto/077436f9ad6d43e55bea9b831638b8c4b733aa26
[4]https://ieeexplore.ieee.org/document/114797
[5]https://news.harvard.edu/gazette/story/2016/08/the-first-autonomous-entirely-soft-robot/
[6]https://news.cornell.edu/stories/2022/12/soft-robot-detects-damage-and-heals-itself
[7]https://www.wevolver.com/article/powering-soft-robotics-a-deeper-look-at-soft-robotics-actuators
[8]https://www.elveflow.com/microfluidic-reviews/general-microfluidics/soft-robot/
[9]https://dash.harvard.edu/bitstream/handle/1/12388526/82522713.pdf
[10]https://www.elveflow.com/microfluidic-reviews/general-microfluidics/soft-robot
[11]https://soft-gripping.com/discover/soft-robotics-vs-hard-robotics/
[12]https://doi.org/10.1080/24725838.2023.2284193
[13]https://news.stanford.edu/2015/10/15/artificial-skin-bao-101515/
[14]https://www.iotworldtoday.com/robotics/self-healing-robot-unveiled-
[15]https://www.science.org/doi/10.1126/scirobotics.adi5908
[16]https://www.cnbc.com/2020/03/22/how-wearable-robots-are-helping-people-with-paralysis-walk-again.html
[17]https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9885398/
[18]https://iopscience.iop.org/article/10.1088/2516-1091/acb57a
[19]https://onlinelibrary.wiley.com/doi/10.1002/aisy.202100254
[20]https://ieeexplore.ieee.org/abstract/document/9981652
[21]https://onlinelibrary.wiley.com/doi/10.1002/aisy.202200326
[22]https://link.springer.com/chapter/10.1007/978-981-16-5180-9_13
[23]https://www.genengnews.com/topics/cancer/tiny-robots-detect-and-treat-cancer-by-traveling-deep-into-the-lungs/
[24]https://wyss.harvard.edu/technology/soft-exosuits-for-lower-extremity-mobility/
[25]https://thereader.mitpress.mit.edu/the-plant-inspired-robots-that-could-colonize-mars/
[26]https://www.frontiersin.org/articles/10.3389/frobt.2023.1129827/full

About the Author

Brandon has been a deep tech journalist, storyteller, and technical writer for more than a decade, covering software startups, semiconductor giants, and everything in between. His focus areas include embedded processors, hardware, software, and tools as they relate to electronic system integration, IoT/industry 4.0 deployments, and edge AI use cases. He is also an accomplished podcaster, YouTuber, event moderator, and conference presenter, and has held roles as editor-in-chief and technology editor at various electronics engineering trade publications. When not inspiring large B2B tech audiences to action, Brandon coaches Phoenix-area sports franchises through the TV.

Profile Photo of Brandon Lewis