Recreating skin in electronic form is a huge challenge for researchers and engineers, but the benefits for humans (and machines) are enormous.
Recreating a human being as a robot is no easy task. Making a human brain? Researchers have already started to model synapses and neurons in software and hardware. Develop a robot capable of moving like a human? Researchers are already building artificial muscles, joints, and tendons for bipedal machines.
But arguably one of the biggest challenges in building a truly human robot lies in our skin. First of all, there is the question of size: at 1.5 to 2 meters across, the skin is the largest organ in the human body. Then there's everything she does. In addition to keeping our interior in and the outside world out, the skin have separate receptors to sense different sensations: pressure, texture, vibration, cold and heat, as well as the ability to capture various sensations at the slightest touch, all over the body.
For anyone looking to create a robot skin capable of capturing the same number of sensations as its human declination or adapting like biological skin and collecting and processing information from millions of sensors per second, one of the main obstacles to overcome is power. The tissues of the skin contain millions of receptors which collect information. A robot with a similar density of sensors, sampling this information hundreds or thousands of times per second, would require a lot of energy and processing power.
Insane computing power
This is why, after covering a single robotic arm with electronic skin and processing the data using traditional computer methods, Gordon Cheng, professor of cognitive systems at the Technical University of Munich, was convinced that the systems used by the human body would be a much more useful model.
“When we used conventional wisdom and power to make sense of the data, it worked to a certain extent. But when we tried to scale up, we needed more and more computers, ”says Gordon Cheng. One of the smartest things about the whole biological system is that it doesn't send information to the brain until something changes and it isn't needed, he explains.
This is because the skin is designed to transmit only the information the brain needs, when it needs it. When you put on your socks this morning, your skin told your brain that they were covering your feet. But your skin knows that your brain doesn't need to be constantly informed that you are still wearing socks all day long. The skin receptors therefore increase the signal when the sock is put on, and decrease it until it is taken off at the end of the day.
Create skin "cells"
Cheng's lab created skin "cells" with sensors for movement, pressure, and other sensations, which only transmit information when a change occurs. The event-based system reduces power consumption by 90%, making their widespread use more achievable.
The University of Munich lab used the cells to cover most of a human-sized robot called the H-1, which can use feedback from cells to adjust its movements: the cells on its arms. help determine the right pressure to use for a hug, while cells on the soles of her feet help her adjust to walking on different terrains.
At the National University of Singapore, researchers also aim to make the skin less restrictive for robotic systems by using biologically-inspired computing: the electronic skin of the Singapore University uses neuromorphic chips, which are inspired by of the way information is processed in the human brain, to limit the energy requirements of the system.
Intel in the loop
The National University of Singapore skin, which uses Intel's Loihi neuromorphic chips, is also event driven. It is modeled on the “peaks” of activity that are transmitted by human nerve fibers and only transmits information when there is a change in the sensations it receives. This not only reduces the amount of data, but also consumes 100 times less energy.
If the human skin and nervous system can serve as a model for electronic equivalents, the biological model goes no further. Our skin, brain and nerves are not updated: we have to maintain roughly the same processing power and the same sensing abilities throughout our lives. However, thanks to advances in software and hardware, the capabilities of the skin of robots will eventually exceed those of human skin.
The skin at the National University of Singapore already senses touch more than 1,000 times faster than its human counterpart, and the capabilities of electronic skin will only improve over time. "We have already demonstrated that, through the use of our technology, we are able to impart not only the sense of touch, but also a superhuman sense of touch," says Benjamin CK Tee, deputy director of the extension service and innovation from the Department of Materials Science and Engineering at the National University of Singapore.
A breakthrough for our health?
Like human skin, the skin of robots will need to feel pain, in order to serve as an early warning system that warns the robot when it is in danger of being damaged. Australian University RMIT has created a prototype robotic skin capable of sensing pain, realistically recreating how the skin constantly senses sensations such as heat or cold, but pain is only registered when certain thresholds are crossed: when the heat becomes strong enough to damage the skin, for example.
The sensitivity of the electronic skin could be adjusted to recreate other skin conditions, such as sunburn. And recreating these conditions in the skin of a robot could help researchers working on the biological versions better understand and treat them.
This isn't the only way a fully realized electronic skin could help humans. Current prostheses may look like human joints and even move the same, but they do not have the same sensing capabilities. “A prosthetic arm can dramatically improve people's lives, but it's not yet quite close to a human limb. He doesn't have the capacity to smell. Imagine that this electronic skin extends over your prosthetic arm, it could bring it a little closer to a real human member ”, explains Madhu Bhaskaran, professor and co-director of the research group on functional materials and Microsystems at the RMIT university.
Resist to last better
Sensory signals from the skin are already transmitted to the human brain by electrical signals. In theory, connecting the electronics of prosthesis to the nervous system shouldn't require a lot of additional technical knowledge.
On the other hand, it will require additional knowledge of materials science. Any electronic skin that connects to human tissue will need to be biocompatible (i.e. the body will not try to reject it) and able to withstand the harsh environment of the human body (the salty and humid environment). Tissue is generally not a welcoming place for electronics).
It will also have to withstand all the stretching and bending that human limbs undergo, without cracking or deforming, and last a long time. Human skin also has a pretty impressive self-healing ability: a small cut can go away in a matter of days, and for larger wounds it can create a whole new material to cover the space, in the form of a scar.
Researchers are already working on materials with self-healing properties similar to those of human skin. Engineers at Carnegie Mellon University, for example, have created a class of flexible, stretchable polymers that contain liquid alloys that allow them to self-heal when punctured, for example. Other researchers have suggested that graphene could also be used to create self-healing robot skin.
At the same time, researchers at the National University of Singapore have developed a foam material, with nerve-like electrodes embedded in it, which can self-healing if damaged. While many engineering and material issues - from longevity to biocompatibility and even aesthetics - remain to be addressed, the benefits of electronic skin for robots and humans are evident.
“The skin gives us a whole sense of the world. In addition to that, it gives us the context to interact with others, for example with a handshake, or a punch, ”says Benjamin CK Tee, of the Singaporean University. “I think the technologies we are developing will allow robots and humans to actually collaborate much more effectively and the social impacts can be very positive. "