Bringing soft robotics into the medical world

What are soft robotics?

A soft robot is composed of soft materials and actuation methods, comprising of elastomers primarily driven by compressed air (pneumatics) of some form. The first soft robot generally accepted was the pneumatic artificial muscle (PAM), which was invented in the 1950s to act as a reinforcement muscle  to aid polio patients suffering from paralysis. PAM comprised of an elastomeric material, actuated with pressurised air to contract and extend a pneumatic bladder. This technology was used due to it being lighter and easier to manipulate than actuators at the time, a factor that still upholds as the greatest strength of soft robotics today.

Soft robotics are largely considered to be a form of bio-inspired engineering. The octopus has often been a point of inspiration in this space, bearing a close resemblance to many soft robots that have been prototyped to date. Research into creating a soft robotic octopus has been tackled by several academic institutions, such as Harvard’s “Octobot”, which was claimed to be the first entirely autonomous soft robot that contains no rigid components, in addition to a tentacle robot which can gently grasp fragile objects.

Although soft robotics was first utilised 70 years ago and are now widely accepted as a form of robotics in academia, it has not yet managed to break into the world commercially. However, the research in academia is starting to mature, use applications are growing and it is now only a matter of time before such technology makes it to market.

What can soft robotics be used for?

Soft robotics are adaptable for use in a wide range of tasks in different environments, due to their flexibility and ability for complex interactions. Research to date has explored various possible applications of soft robotics. For example, we spoke to Melanie Simons, who completed a PhD in soft robotics and has conducted research into soft smart wearable prosthetic gloves. Simons’ colleagues have also been conducting research on anxiety-easing pillows, suggesting potential for soft robotics in both of these areas in the future.

Soft robots are generally considered safe and suitable for human contact, owing to their ability to ‘comply’ and deform around an object of interest. Where rigid robotics need to be controlled to extreme precision, a soft robot could overshoot and still achieve its overall goal without compromising safety, owing to the fact it can autonomously conform its shape to different objects. Several academics are currently paving the way for what this might look like for a user – watch this space.

One of the key benefits of soft robotics is that their lightweight and portable form allows them to access hard-to-reach locations. This means there could be numerous use cases for soft robotics in medical applications and in healthcare, a current example being Nanoflex Robotics, which is advancing medical robotic interventions for the treatment of ischemic stroke within the brain.

The following are some other notable examples of soft robot applications being developed for use in the medical space:

Active othotic trousers

“The Right Trouser” project, being run by the University of Bristol and University of the West of England, demonstrates a wearable device that assists older people and people with disabilities to live more independently and improve their quality of life. The trousers are based on soft robotic technology and assist a range of motion for ease of walking, dressing, undressing, standing up, sitting down and climbing stairs. The idea behind them is to help older people stay independent for longer, potentially reducing accidents and costs on healthcare and care homes.

“The Right Trouser” soft robot technology works by contracting very small soft air pockets. The wearable device uses smart materials that can bend, twist or contract when electrically stimulated. While these show some promise, the current challenge is to make them smaller, thinner and lighter, to hide within someone’s clothing.

Trousers are not the only potential application for this technology either. Other research has been exploring artificial e-skins that are composed of a pressure sensitive rubber and a grid of organic field effect transistors. This could result in more mass production wearable devices, for example rehabilitation gloves to aid in dexterity after injury or illness, such as a stroke.

Surgical tools

Other research into the potential applications of soft robotics include soft endoscopes for next-generation gastrointestinal surgery. A self-propelled endoscopic robot being developed by Imperial College London actively changes its stiffness and shape, allowing it to propel itself into difficult conditions such as the human gastrointestinal tract. This could offer a great alternative to traditional endoscopic instruments, which can cause patient pain due to deforming or perforating tissue. A soft robot, comparatively, can match the soft tissue of the gastrointestinal tract to create an inherently safer device. However, it should be noted that this is only part of the challenge with surgical instruments. Pain is also caused by positioning, dexterity, force exertion and visualisation.

These research examples demonstrate the need to bring together the controllability of rigid robotics, the access capabilities of flexible instruments and the safety of soft materials, to solve different healthcare problems. Although soft robotics is a heavily researched subject, it’s clear there could be a real use for this technology if these devices can be produced commercially and safely.

The challenges of commercialising soft robotics

The clear challenge of any novel academic research is converting that knowledge into something commercially viable. The current soft robotics in development are being produced on a small scale with innovative technology. Sourcing manufacturers and resources, meeting regulatory requirements and ensuring success in the operational environment are just a few of the biggest hurdles that must be overcome before soft robot applications can be taken from research to development and finally deployment.

One of the biggest technical challenges of soft robotics is that they are difficult to control due to their many degrees of freedom. They are also less powerful and precise than rigid robotics and today’s electronics are not adapted, which means support equipment can limit the motion and autonomy available to them. Generally, the core challenges facing soft robotics can be put into two key categories – technical challenges and market acceptance.

Technical challenges

• Actuation

  Soft robots currently require extremely high voltages to operate. Developing efficient and reliable actuation and power systems capable of generating required forces for soft robotic movements will be a key hurdle in bringing them into commercial circulation. To put this in perspective, the batteries that are currently being used to power soft robots have an energy density that is 10 – 100 times lower than the sugars and fats that power natural muscles, which is largely what soft robotics is founded on.

  Soft actuators and sensors are generally not made on a commercial scale at this time, meaning further development will be required before soft robotics can be progressed from a manufacturing perspective.

• Manufacture and cost efficiency

  Developing scalable and cost-effective manufacturing processes that ensure consistency is an unknown hurdle and therefore presents significant risk. As with any medical device, this will need to be considered during the development of soft robots to ensure commercial viability.

  Soft robots are currently claimed in academia to be low cost. However, since they have not yet been manufactured commercially, this benefit is currently difficult to quantify. Companies looking to innovate in this space will likely be treading new ground when it comes to determining cost efficiency for manufacturing.

• Materials development

  Soft robotic materials are generally considered to be biocompatible and environmentally adaptable. However, the impact of different conditions – including temperature, humidity and preconditioning – to ensure robust performance must be proven. Currently, soft robotics are widely demonstrated to be capable within a laboratory. The question is, how does this transfer into a true-to-use environment?

  Another challenge is that no current material has all the properties to match human muscle fibres. The material requirements for soft robots include enduring high stress and strains, a high energy density and the need to operate with many cycles. New and innovative materials are regularly being discovered to enable the ‘soft actuation’ required in soft robotics. Shape memory alloys, for example, offer possibilities for the development of novel sensors and actuators and fall into the category of ‘smart materials’. These materials change shape when receiving a stimulus, due to the phase transformation of the materials and are a key enabler for smart robotic technology.

  A great example of an innovative smart material with potential applications in soft robotics is spider silk, which was first cloned in 1990 in an attempt to see if it could be incorporated into other organisms to produce the silk. The goal in doing so was to commercially manufacture the silk, due to its strong material despite its light weight. In 2019, this spider silk material was found to contract and twist in humid conditions, suggesting it could be used to perform activities such as controlling a valve or sensors when entering a humid or water-based solution. As more of these innovative materials emerge, there will be greater capabilities for soft robotics.

• Control, sensing, and complex motion

  Soft sensors play a significant part in soft robotics and present another challenge for their development. Currently, most sensors within rigid robotics do not conform to shape; however, recent rapid advancements of artificial skin and flexible electronics are starting to enable soft sensing as a possibility. This is shown by a wide range of research into resistive, capacitive, triboelectric, magnetic and optical sensors (amongst others) that are exhibiting excellent sensing abilities.

  Smart soft sensing has lagged behind the development of soft actuation. To scale these sensors up to full product development is a complex task on its own. Doing so could help to overcome other issues, however. Soft robotics generally form nonlinear motions, resulting in the complex challenge of overcoming kinematics and path planning for compliant structures. Soft sensing and AI may pave the way to overcome this.

Market acceptance

• Soft robotics regulations

  There are strict regulations on anything novel in the medical market. Regulatory standards for ethical approval on human studies is thorough and navigating this process can be time-consuming and costly for any company looking to innovate in this space.

  Most medical devices also require sterilisation, which is a particular challenge with soft materials. Meanwhile, the impact by medical products on the environment is likely going to become regulated sometime in the near future. Soft materials are generally not very successful with reuse and recycling, meaning sustainability may pose another challenge for soft robotic manufacturers.

• Integration and training

  There have been claims that less training is required for a soft robot due to the use of a compliant material. Significant training is required for any new technology, especially within surgery where any mistakes or malfunctions could be fatal. In fact, some consider training the surgeon is more important than the safety of the robot itself, since adding a robot to the surgical equation generally creates another potential entry point for error.

  Alongside training, many potential users and recipients will have to be convinced of the capabilities and benefits of soft robotics before widespread adoption can be achieved.

• Lack of standardisation and limited commercially available components

  Due to soft robotics being a relatively new field, the current lack of standardised design and process can make it difficult for businesses to integrate and develop soft robotic solutions.

Why are soft robotics now gaining traction?

Despite these challenges, due to technological developments, soft robotics is starting to gain traction. This ‘soft’ introduction to market is due to several reasons:

Increased computing power

Computers have become smaller and more powerful. The use of artificial intelligence, machine learning and data-driven control now have the ability to improve the precision and accuracy of soft robotics. Research by Professor Kaspar Althoefer, Queen Mary University of London, has highlighted that advancements in computer power could even see surgeons operating soft robots without extensive training.

Materials and manufacturing developments

Materials, manufacture and soft composite technologies have also improved. Elastomers are now capable of being easier to sterilise, biocompatible and transparent. Recent advancements in soft lithography, microfluidics and 4D printing have also helped to progress soft robotics technology. 4D printing is different to 3D printing due its use of ‘smart materials’ and shape memory alloys. These materials come in different forms to create actuation, which deform in a predictable way under a stimulus, such as a change in light (photosensitive), electric field (dielectric), heat (combustion) or pH level.

The advancement in soft composites aid the generation of motion with these types of materials and are controlled by combining a softer material with a stiffer or inextensible layer, which can enable the actuation method to deform in a particular direction and orientation.