Smart foam gives robots a human-like sense of touch

Robots and machines are getting smarter with the advancement of artificial intelligence, but they still lack the ability to touch and feel their subtle and complex surroundings like human beings. Now researchers from the National University of Singapore (NUS) have invented a smart foam that can quite literally give machines more of a human touch, describing their results in the journal Nature Communications.

Equipping robots with human senses

The human sense of touch enables people to manipulate objects and operate effectively in unfamiliar environments. When machines that interact with humans possess this capability, robotic motion can be smoother, safer and more predictable.

Take machines such as cleaning robots and robotic waiters as an example. If they can detect human presence from a distance and use the information to deduce human intentions, they can react faster and avoid colliding with people, especially in crowded areas.

While some advanced electronic skins could sense pressure when they come into direct contact with an object, none is able to sense the direction of movement of adjacent objects as this requires more complex sensing abilities. NUS researchers, led by Assistant Professor Benjamin Tee, decided to find a solution to address this technological challenge.

AiFoam

Artificially innervated foam (AiFoam) is a highly elastic polymer created by mixing a Teflon-like substance, known as a fluoropolymer, with a surfactant that lowers surface tension. The result is a soft and spongy material that, when separated into pieces, fuses easily back into one piece. The team then infused the material with microscopic metal particles, which enables the foam to sense the presence of a capacitive object like a human finger.

The foam can be made more sensitive when compared to conventional thin sheets, and its softness can also be better controlled by changing the air-to-material ratio. It is able to better sense the proximity of human presence compared with other commonly used e-skin materials such as silicones; in fact, AiFoam can detect the presence of human fingers from centimetres away. Its formulation also enables it to more efficiently detect human presence — unlike other proximity sensors that typically make use of light and reflections, which can lead to higher false positive or negative rates.

“By having a foam with our special formulation, we can sense both pressure and proximity easily,” Asst Prof Tee said.

To mimic the sensitive nerve endings in the human skin, the researchers embedded fine cylinder-shaped electrodes underneath the surface of the foam. They are able to detect the direction of the applied force, not just the amount of force. This would enable robots to understand human intentions better, or know that an object in contact is about to slip, so that they can react more quickly and appropriately.

The surfactant within the foam enables it to self-heal even after being cut, making it ideal for use in applications like prosthetics. The foam can be stretched to more than twice its length — by 230% — without breaking. To heal cuts, the researchers heated it to 70°C over four days. The material healed about 70%, and could still be stretched to almost twice its length (180%).

Future applications

AiFoam is claimed to be the first smart foam in the world that can simultaneously mimic the human sense of touch, sense nearby objects without actually touching and repair itself when damaged, which would potentially make robots more intelligent and interactive. The NUS team is now exploring partnerships with robotics companies, as well as looking to expand the use of AiFoam to areas like robotics and prosthetics, to enable more seamless interactions between humans and machines.

“The self-healing properties of AiFoam, as well as its sensitivity to a contact force, will enable robots to better judge human intention, and build a future generation of robots that will react to environmental changes quickly and more efficiently,” Asst Prof Tee concluded.

Image credit: ©Vladislav Ociacia/Dollar Photo Club

Please follow us and share on Twitter and Facebook. You can also subscribe for FREE to our weekly newsletter and bimonthly magazine.