Cookies   I display ads to cover the expenses. See the privacy policy for more information. You can keep or reject the ads.

Video thumbnail
Traditionally when we think about robots we think about rigid pieces
that are linked together. Things like C3PO. Things like WALL-E.
What we're trying to do is build robots that emulate biological systems.
So they have soft components, they are very deformable, they can
squeeze through small spaces and they can interact safely with people.
And so what we're trying to do is to go away from metal a little bit.
We wanted to develop tunable stiffness structures and materials.
The idea is that the robot should be soft in situations where we want to
conform to the environment or squeeze through tight spaces, but we
also would need it to be rigid when we need to apply loads on the
environment so that we can adequately push against objects and
that sort of thing.
So what we were looking for was a material that could shift between soft
and hard states. So what you're looking at here is a soft scaffold of foam
that has been coated in wax. When the wax is heated you get the soft
structure and when the wax re-solidifies it regains its rigidity.
So imagine if your components were more compliant then perhaps we
could increase the robots capabilities. For example if you wanted to use
this composite foam-wax that has tunable stiffness properties to make
some autonomous robot or to control the shape of a robot you can have
different segments of the foam coated in wax and selectively change the
stiffness of them. So lets say you have three segments and you wanted
to just bend the middle one, you keep the outer two segments rigid
in the cold state, and you heat up the middle one so that it softens and
you can bend it; for example by pulling a string or a cable that goes
down the length just to bend that one. And then of course if you want
that shape or that segment to freeze in its new shape, you let it cool in
whatever bent position, at room temperature.
I think that the structures you are seeing here are just the beginning of
a whole new class of robotics. Again, imagine robots that have the
same capabilities of biological systems. Mice, we all know we can't
keep mice out of anything because they can squeeze through tiny, tiny
cracks. This is something that could be useful for robots in
search-and-rescue applications, where you have to go through rubble.
It could be useful in medical applications where you have to squeeze
through small parts of the body. It could be useful in areas where robots
have to interact safely with humans; there's a whole host of applications
out there.