Baymax-Like "Inflatable" Smart Hand Outperforms Rigid Neuroprosthetics at a Fraction of the Cost

Engineers at the Massachusetts Institute of Technology (MIT) and Shanghai Jiao Tong University have designed an inflatable robotic hand, designed to offer amputees real-time tactile control at a fraction of the cost of commercial equivalents.

Known as "neuroprosthetics," articulated bionic limbs are helping amputees by offering the ability to trigger a range of movements through residual muscle motion - effectively giving them a limb which they can control much like the original. The only problem: The devices are typically rigid, heavy, and extremely expensive.

That's where a team of researchers in the US and Shanghai come in, with a design which uses a soft inflatable system in place of a rigid metal skeleton — and costs just a fraction as much as commercial equivalents to build.

“This is not a product yet, but the performance is already similar or superior to existing neuroprosthetics, which we’re excited about," says Xuanhe Zhao, professor of mechanical engineering and of civil and environmental engineering at MIT, of the work. "There’s huge potential to make this soft prosthetic very low cost, for low-income families who have suffered from amputation."

Built from around $500 in components, the prototype "smart hand" was run through a range of tasks — from zipping up a suitcase and pouring a carton of fruit juice to petting a cat — and was rated as equivalent to or better than rigid versions.

Better still, the prosthetic offers tactile feedback — reproducing what the team calls "primitive sensation" in the patients' residual limb — and proved durable, even in the face of hammer blows or being run over by a car.

Built from a stretch elastomer connected to a 3D-printed palm, the five-fingered hand bears a similarity to Baymax from Disney's Big Hero 6 — and works in much the same way, inflating and deflating pneumatics in place of traditional motors. Electromyographic (EMG) sensors connected to an Arduino Nano turn muscle signals into one of four movements — but this could be extended.

"We now have four grasp types. There can be more, claims Zhao. "This design can be improved, with better decoding technology, higher-density myoelectric arrays, and a more compact pump that could be worn on the wrist. We also want to customize the design for mass production, so we can translate soft robotic technology to benefit society."

The team's work has been published under closed-access terms in the journal Nature Biomedical Engineering.