Robots are becoming more dextrous than ever. This incredibly detailed robotic hand is one of the most accurate mechanical devices modelled on the human body that we’ve ever seen — because it simply swaps bones and tissue for exact mechanical replicas.
Built by researchers from the University of Washington, the hand is modelled on that of a real human: its skeleton is actually just that, a 3D-printed version of all the bones in a hand created by laser-scanning an actual human being. Those complex pieces are held together with a series of artificial ligaments, replicated by high strength Spectra strings, with laser-cut latex sheets to replicate the soft tissue which makes your joints nicely compliant.
Tendons are also replicated using more Spectra springs, which are sheathed in yet more latex sheets, just as soft tissue surrounds a normal tendon. There’s also a complex laser-cut web of latex used to mimic the extensor hood, which is a a strange muscular structure that wraps around the fingers and helps you create torque with your hand. Then, finally, the muscles are replicated using 10 Dynamixel servo motors.
Phew. Well, we did say it was complex.
And it works amazingly well. The fact that the hand’s so anatomically correct provides it with an impressive range of motion when controlled remotely.
It can also grasp a wide range of objects very easily, too.
Perhaps most impressive, though, is that the researchers think that it could be used not just as a an artificial hand, but as an actual scaffold for growing a new one. Zhe Xu, one of the researchers, explains to IEEE Spectrum:
Biocompatible materials can now be printed to form bone structures, biodegradable artificial ligaments have been used to replace the torn anterior cruciate ligaments, human muscles have been successfully cultivated inside petri dish, and peripheral nerves can also be regenerated given the right conditions. All of the these promising technologies require suitable scaffolds for the growth of grafted cells. We are going to collaborate with researchers from biology and tissue engineering to further explore its potential to serve as a bio-fabricated device/scaffold in the emerging fields of neuroprosthetics and limb regeneration.
That’s a bold claim, but not a ridiculous one. [IEEE Spectrum]