9:59:59.000,9:59:59.000
So robots.

9:59:59.000,9:59:59.000
Robots can be programmed[br]to do the same task millions of times

9:59:59.000,9:59:59.000
with minimal error, something[br]very difficult for us, right?

9:59:59.000,9:59:59.000
And it can be very impressive[br]to watch them at work.

9:59:59.000,9:59:59.000
Look at them.

9:59:59.000,9:59:59.000
I could watch them for hours.

9:59:59.000,9:59:59.000
No?

9:59:59.000,9:59:59.000
But what is less impressive[br]that if you take this robot

9:59:59.000,9:59:59.000
out of the factories

9:59:59.000,9:59:59.000
where the environments are not[br]perfectly known and measured like here,

9:59:59.000,9:59:59.000
to do even a simple task[br]which doesn't require much precision,

9:59:59.000,9:59:59.000
and this is what can happen.

9:59:59.000,9:59:59.000
I mean, opening a door,[br]you don't require much precision.

9:59:59.000,9:59:59.000
(Laughter)

9:59:59.000,9:59:59.000
Or, a small error in the measurements,

9:59:59.000,9:59:59.000
you miss the ?? and that's it

9:59:59.000,9:59:59.000
(Laughter)

9:59:59.000,9:59:59.000
with no way of recovering[br]most of the time.

9:59:59.000,9:59:59.000
So why is that?

9:59:59.000,9:59:59.000
Well, for many years,

9:59:59.000,9:59:59.000
robots have been designed[br]to emphasize speed and precision,

9:59:59.000,9:59:59.000
and this translates[br]in a very specific architecture.

9:59:59.000,9:59:59.000
If we take a robot term,

9:59:59.000,9:59:59.000
it's a very well-defined[br]set of rigid links

9:59:59.000,9:59:59.000
and mortars who are called actuators,

9:59:59.000,9:59:59.000
they move the links above the joins.

9:59:59.000,9:59:59.000
In this ?? structure,

9:59:59.000,9:59:59.000
you have to perfectly[br]measure your environment,

9:59:59.000,9:59:59.000
so what is around,

9:59:59.000,9:59:59.000
and you have to perfectly[br]program every movement

9:59:59.000,9:59:59.000
of the robot joints,

9:59:59.000,9:59:59.000
because a small error[br]can generate a very large fault,

9:59:59.000,9:59:59.000
so you can damage something[br]or you can get your robot damaged

9:59:59.000,9:59:59.000
if something is harder.

9:59:59.000,9:59:59.000
So let's talk about them a moment,

9:59:59.000,9:59:59.000
and don't think about[br]the brains of these robots

9:59:59.000,9:59:59.000
or how carefully we program them,

9:59:59.000,9:59:59.000
but rather look at their bodies.

9:59:59.000,9:59:59.000
There is obviously[br]something wrong with it,

9:59:59.000,9:59:59.000
because what makes a robot[br]precise and strong

9:59:59.000,9:59:59.000
also makes them ridiculously dangerous[br]and ineffective in the real world,

9:59:59.000,9:59:59.000
because their body cannot deform

9:59:59.000,9:59:59.000
or better adjust to the interaction[br]with the real world.

9:59:59.000,9:59:59.000
So think about the opposite approach,

9:59:59.000,9:59:59.000
being softer than[br]anything else around you.

9:59:59.000,9:59:59.000
Well, maybe you think that you're not[br]really able to do anything if you're soft,

9:59:59.000,9:59:59.000
probably.

9:59:59.000,9:59:59.000
Well, nature teaches us the opposite.

9:59:59.000,9:59:59.000
For example, at the bottom of the ocean[br]under thousands of pounds

9:59:59.000,9:59:59.000
of ?? pressure,

9:59:59.000,9:59:59.000
a completely soft animal

9:59:59.000,9:59:59.000
can move and interact with a much[br]stiffer object than him.

9:59:59.000,9:59:59.000
He works by carrying around[br]this coconut shell

9:59:59.000,9:59:59.000
thanks to the flexibility[br]of his tentacles,

9:59:59.000,9:59:59.000
which serve as both his feet and hands.

9:59:59.000,9:59:59.000
And apparently, an octopus[br]can also open a jar.

9:59:59.000,9:59:59.000
It's pretty impressive, right?

9:59:59.000,9:59:59.000
But clearly, this is not enabled[br]just by the brain of this animal,

9:59:59.000,9:59:59.000
but also by his body,

9:59:59.000,9:59:59.000
and it's a clear example,[br]maybe the clearest example,

9:59:59.000,9:59:59.000
of embodied intelligence,

9:59:59.000,9:59:59.000
which is a kind of intelligence[br]that all living organisms have.

9:59:59.000,9:59:59.000
We all have that.

9:59:59.000,9:59:59.000
Our body, its shape,[br]material and structure,

9:59:59.000,9:59:59.000
plays a fundamental role[br]during a physical task,

9:59:59.000,9:59:59.000
because we can conform[br]to our environment

9:59:59.000,9:59:59.000
so we can succeed in a large[br]variety of situations

9:59:59.000,9:59:59.000
without much planning[br]or calculations ahead.

9:59:59.000,9:59:59.000
So why don't we put[br]some of this embodied intelligence

9:59:59.000,9:59:59.000
into our robotic machines

9:59:59.000,9:59:59.000
to release them from relying[br]on excessive work

9:59:59.000,9:59:59.000
on computation and sensing?

9:59:59.000,9:59:59.000
Well, to do that we can follow[br]the strategy of nature,

9:59:59.000,9:59:59.000
because with evolution,[br]she's done a pretty good job

9:59:59.000,9:59:59.000
in designing machines[br]for environment interaction,

9:59:59.000,9:59:59.000
and it's easy to notice that nature[br]uses soft material frequently

9:59:59.000,9:59:59.000
and stiff material sparingly.

9:59:59.000,9:59:59.000
And this is what is done[br]in this new field or robotics

9:59:59.000,9:59:59.000
which is called soft robotics,

9:59:59.000,9:59:59.000
in which the main objective[br]is not to make super-precise machines

9:59:59.000,9:59:59.000
because we've already got them,

9:59:59.000,9:59:59.000
but to make robots able to face[br]unexpected situations in the real world,

9:59:59.000,9:59:59.000
so able to go out there.

9:59:59.000,9:59:59.000
And what makes a robot soft[br]is first of all his compliant body,

9:59:59.000,9:59:59.000
which is made of materials or structures[br]that can undergo very large deformations,

9:59:59.000,9:59:59.000
so no more rigid links,

9:59:59.000,9:59:59.000
and secondly to move them[br]we use what we call distributed actuation,

9:59:59.000,9:59:59.000
so we have to control continuously[br]the shape of this very deformable body,

9:59:59.000,9:59:59.000
which is the fact of having[br]a lot of links and joints,

9:59:59.000,9:59:59.000
but we don't have[br]any stiff structure at all.

9:59:59.000,9:59:59.000
So you can imagine that building[br]a soft robot is a very different process

9:59:59.000,9:59:59.000
than stiff robotics, where[br]you have links, gears, screws

9:59:59.000,9:59:59.000
that you must combine[br]in a very defined way.

9:59:59.000,9:59:59.000
In soft robots, you just build[br]your actuator from scratch

9:59:59.000,9:59:59.000
most of the time,

9:59:59.000,9:59:59.000
but you shape your flexible material

9:59:59.000,9:59:59.000
to the form that responds[br]to a certain input.

9:59:59.000,9:59:59.000
For example here, you can just[br]deform a structure

9:59:59.000,9:59:59.000
doing a fairly complex shape

9:59:59.000,9:59:59.000
if you think about doing the same[br]with rigid links and joints,

9:59:59.000,9:59:59.000
and here what you use is just one input,

9:59:59.000,9:59:59.000
such as air pressure.

9:59:59.000,9:59:59.000
Okay, but let's see[br]some cool examples of soft robots.

9:59:59.000,9:59:59.000
Here is a little cute guy[br]developed by Harvard University,

9:59:59.000,9:59:59.000
and he works thanks to waves[br]of pressure applied along its body,

9:59:59.000,9:59:59.000
and thanks to the flexibility he can[br]also sneak under a low bridge,

9:59:59.000,9:59:59.000
keep walking,

9:59:59.000,9:59:59.000
and then keep walking[br]a little bit different afterwards.

9:59:59.000,9:59:59.000
And it's a very preliminary prototype,

9:59:59.000,9:59:59.000
but they also built a more robust version

9:59:59.000,9:59:59.000
with power on board that can actually[br]be sent out in the world

9:59:59.000,9:59:59.000
and face real-world interactions

9:59:59.000,9:59:59.000
like a car passing it over it,

9:59:59.000,9:59:59.000
and keep working.

9:59:59.000,9:59:59.000
(Laughter)

9:59:59.000,9:59:59.000
It's cute.

9:59:59.000,9:59:59.000
(Laughter)

9:59:59.000,9:59:59.000
Or a robotic fish which swims[br]like a real fish does

9:59:59.000,9:59:59.000
in water simply because it has a soft tail[br]with distributed actuation

9:59:59.000,9:59:59.000
using still air pressure.

9:59:59.000,9:59:59.000
That was from MIT,

9:59:59.000,9:59:59.000
and of course we have a robotic octopus.

9:59:59.000,9:59:59.000
This was actually one of[br]the first projects developed

9:59:59.000,9:59:59.000
in this new field of soft robots.

9:59:59.000,9:59:59.000
Here you see the artificial tentacle,

9:59:59.000,9:59:59.000
but they actually built and entire machine

9:59:59.000,9:59:59.000
with several tentacles they could[br]just throw in the water,

9:59:59.000,9:59:59.000
and you see that it can kind of go around[br]and do submarine exploration

9:59:59.000,9:59:59.000
in a different way[br]than rigid robots would do.

9:59:59.000,9:59:59.000
But this is very important for delicate[br]environments such as coral reefs.

9:59:59.000,9:59:59.000
Let's go back to the ground.

9:59:59.000,9:59:59.000
Here you see the view

9:59:59.000,9:59:59.000
from a growing robot developed[br]by my colleagues in Stanford.

9:59:59.000,9:59:59.000
You see the camera fixed on top.

9:59:59.000,9:59:59.000
And this robot is particular[br]because using air pressure

9:59:59.000,9:59:59.000
it grows from the tip[br]while the rest of the body

9:59:59.000,9:59:59.000
stays in firm contact[br]with the environment.

9:59:59.000,9:59:59.000
And this is inspired[br]by plants, not animals,

9:59:59.000,9:59:59.000
which grows via the material[br]in a similar manner

9:59:59.000,9:59:59.000
so it can face a pretty large[br]variety of situations.

9:59:59.000,9:59:59.000
But I'm a biomedical engineer,

9:59:59.000,9:59:59.000
and perhaps the application[br]I like the most

9:59:59.000,9:59:59.000
it's in the medical field,

9:59:59.000,9:59:59.000
and it's very difficult to imagine[br]a closer interaction with the human body

9:59:59.000,9:59:59.000
than actually going inside the body.