A quick look at the world’s demographics will reveal a relatively solid truth; the world’s elderly population is growing, and it’s growing faster than any other age group. By the year 2050, the population of individuals age sixty and older is expected to double, and with that kind of growth comes a higher demand for caregivers that can offer assistance not just in nursing homes or hospitals, but around the clock in the comfort of one’s own home.
Robots that can assist with care-giving tasks exist already, but their function is somewhat limited. While they can often ask patients questions about their comfort or monitor their health for accidents or risks, they have one inherent feature that makes them less than ideal for physical care: they’re hard. The world’s robots are made of metal, hard plastics, or equally tough and unyielding materials that aren’t very conducive to the gentle touch that may be required when physically aiding a human being.
However, obstacles are the mother of all invention, and this obstacle is no exception. As material science continues to advance, so does our ability to create different platforms for artificial intelligence to reside in. One such material advance now allows us to create robots with bodies that can shift and reshape when touched, much like human skin and muscle. Naturally, the cost and complexity of producing such robots is inherently prohibitive, but a slightly less recent invention, the 3D printer, may be the key to cheaply mass producing AI robots with this type of material body, thanks to research teams at Purdue University.
By combining a human designed file that dictates how the robot body will move with a computer algorithm that converts it into a 3D model, a humanoid robot with striking dexterity and safety can be made with a conventional 3D printer. The platform’s miniaturized rotors work in tandem with nylon strings to move its limbs in a way that is not all that dissimilar from human muscles and tendons. But unlike the human body, the material of the robot’s body will be capable of stretching compressing up to nine hundred percent of its original length, adding a new realm of functionality.
What this advancement means for medical care is likely that care-giving roles will one day be filled with soft AI robots that can fulfill even the physical care-giving roles usually reserved for nurses. These soft robots may be able to assist with bathing, moving, or getting dressed. They will likely be able to administer a variety of physical medical practices. By 2050, the care-giving market may be filled by more AI robots than it is people.
But what this advancement means for robotics and AI as a whole is something that can only be guessed at. With the ability to adapt their physical bodies to new terrains and situations, these ‘soft’ robots might one day also fill roles in disaster response, search and rescue, or even warfare. And with bodies more reminiscent of humans than any other robotic platform before it, such robots could also feasibly replace customer service roles one day. And while the AI of today is limited to its programming, one can only wonder at the kind of situations a self-learning AI in a very human-like body may present to society.
Videos of Purdue’s most recent breakthroughs with soft robotic technology can be found at their website: https://engineering.purdue.edu/FlexiLab/videos.html