Engineers at Harvard have developed a soft robotic glove that allows people with limited hand mobility to grasp and pick up objects. The device could help the estimated 6.8 million people in the United States who have hand mobility issues, whether from a degenerative condition, stroke, or old age.
Nine patients with ALS, muscular dystrophy, incomplete spinal cord injuries, or complications from a stroke have tested the glove so far.
The goal is to restore independence for people who have lost the ability to grasp, says Conor Walsh, a professor at Harvard’s Wyss Institute for Biologically Inspired Engineering. The project was led by Panagiotis Polygerinos, a technology development fellow in Walsh’s lab. Walsh thinks that within three years the glove will be “suitable for use in the medical environment.”
For hand mobility difficulties, existing robots with hard exoskeletons can act as assistive devices and guide patients through rehabilitation exercises. But a soft robotic glove aligns more flexibly with a patient’s joints, plays nice with soft tissue like human skin, and, since it is much lighter, could eventually be taken home instead of being limited to use in a clinic.
The glove could give patients “the dexterity that they need to perform essential activities of daily life,” says Steve Kelly, president and COO of Myomo, a developer of assistive robotic devices for the arm and hand, who was not involved in the project.
The glove is mechanically programmed to execute a single task, performed with a bending motion of the fingers and a bending and twisting motion of the thumb. The fingers are essentially silicone balloons—pink, rubbery things—with yellow fibers crisscrossed inside. When pressurized water is pumped into the glove from an attached waist pack, the fibers keep the balloon from expanding, so their arrangement programs the finger to bend in a particular way. For example, there are fewer fibers at the knuckles, which induces the finger to bend there.
Polygerinos let me try it out. The outside of the glove is made of a soft neoprene-like fabric, the fingers covered in a wormlike series of clear rubbery rings for grip. I slipped my left hand into the glove, and he flipped the switches. The motor hummed like a belt sander, and without any help from me, my fingers and thumb curled together in a grasping motion. It felt as if someone else’s hand were underneath mine—someone stronger, moving my fingers for me. The glove is customized to fit a patient’s hand so that the joints align properly, and this glove was a little too big for me, but still, it felt comfortable.
“It’s really simple, because all you do is pressurize it and you get this nice complex motion,” says Walsh. “The downside is, it’s that one motion all the time.”
Though that is a limitation, grasping is extremely important and many patients need help with it, says MIT professor Neville Hogan, who creates robots to rehabilitate stroke patients. “Most neurological disorders cause muscle weakness, which leads to impaired grasp strength,” he says. However, stroke patients’ hand muscles are often clenched by default, so Hogan says they often have the most trouble opening their hands. The team says the glove does not currently have enough force to open the hand if the muscles are clenched, but they hope to add that functionality in the future.
They also want to make the device lighter. The glove weighs 10 ounces, and the waist pack containing the battery, controllers, sensors, pump, and water weighs about seven pounds (twice the weight of a 13-inch MacBook Pro).
The glove is operated either by flipping a switch or by voice command. The next step is to design a glove that can move when it detects signals in the patient’s own arm muscles, so that patients can control it more intuitively. Designing such a control system is tricky. Even patients with the same condition have individual variation, and patients have good days and bad days. “So you can go one day, try your electrodes—signals are perfect, you can operate the glove. You go two days later, something is wrong and you don’t get the same signals again,” says Polygerinos.
Kelly thinks the control mechanism will be key. “Whoever has the best control will have the best commercial solution,” he says. “It’s probably reasonable in the five-ish-year time frame to be able to get this as an impaired person,” he estimates.
Asked if he can remember the best thing a patient has said when trying the glove, Polygerinos looks thoughtful, and then his face lights up. “Oh my God, I can pinch again!”
How AI is reinventing what computers are
Three key ways artificial intelligence is changing what it means to compute.
These weird virtual creatures evolve their bodies to solve problems
They show how intelligence and body plans are closely linked—and could unlock AI for robots.
Our brains exist in a state of “controlled hallucination”
Three new books lay bare the weirdness of how our brains process the world around us.
We reviewed three at-home covid tests. The results were mixed.
Over-the-counter coronavirus tests are finally available in the US. Some are more accurate and easier to use than others.
Get the latest updates from
MIT Technology Review
Discover special offers, top stories, upcoming events, and more.