0:00:01.022,0:00:03.472
So robots.

0:00:03.472,0:00:07.415
Robots can be programmed[br]to do the same task millions of times

0:00:07.415,0:00:11.293
with minimal error, something[br]very difficult for us, right?

0:00:11.293,0:00:14.477
And it can be very impressive[br]to watch them at work.

0:00:14.477,0:00:15.756
Look at them.

0:00:15.756,0:00:18.472
I could watch them for hours.

0:00:18.472,0:00:19.622
No?

0:00:19.622,0:00:23.559
But what is less impressive[br]that if you take this robot

0:00:23.559,0:00:24.820
out of the factories

0:00:24.820,0:00:29.223
where the environments are not[br]perfectly known and measured like here,

0:00:29.223,0:00:33.588
to do even a simple task[br]which doesn't require much precision,

0:00:33.588,0:00:35.222
and this is what can happen.

0:00:35.222,0:00:37.512
I mean, opening a door,[br]you don't require much precision.

0:00:37.512,0:00:38.901
(Laughter)

0:00:38.901,0:00:41.706
Or, a small error in the measurements,

0:00:41.706,0:00:43.472
you miss the ?? and that's it

0:00:43.472,0:00:44.765
(Laughter)

0:00:44.765,0:00:47.761
with no way of recovering[br]most of the time.

0:00:47.761,0:00:49.331
So why is that?

0:00:49.331,0:00:51.358
Well, for many years,

0:00:51.358,0:00:54.856
robots have been designed[br]to emphasize speed and precision,

0:00:54.856,0:00:57.828
and this translates[br]in a very specific architecture.

0:00:57.828,0:00:58.874
If we take a robot term,

0:00:58.874,0:01:01.626
it's a very well-defined[br]set of rigid links

0:01:01.626,0:01:03.852
and mortars who are called actuators,

0:01:03.852,0:01:05.920
they move the links above the joins.

0:01:05.920,0:01:07.150
In this ?? structure,

0:01:07.150,0:01:08.790
you have to perfectly[br]measure your environment,

0:01:08.790,0:01:10.802
so what is around,

0:01:10.802,0:01:13.649
and you have to perfectly[br]program every movement

0:01:13.649,0:01:15.274
of the robot joints,

0:01:15.274,0:01:19.158
because a small error[br]can generate a very large fault,

0:01:19.158,0:01:22.194
so you can damage something[br]or you can get your robot damaged

0:01:22.194,0:01:24.307
if something is harder.

0:01:24.307,0:01:26.628
So let's talk about them a moment,

0:01:26.628,0:01:29.397
and don't think about[br]the brains of these robots

0:01:29.397,0:01:32.260
or how carefully we program them,

0:01:32.260,0:01:34.607
but rather look at their bodies.

0:01:34.607,0:01:37.557
There is obviously[br]something wrong with it,

0:01:37.557,0:01:40.972
because what makes a robot[br]precise and strong

0:01:40.972,0:01:45.504
also makes them ridiculously dangerous[br]and ineffective in the real world,

0:01:45.504,0:01:47.572
because their body cannot deform

0:01:47.572,0:01:51.506
or better adjust to the interaction[br]with the real world.

0:01:51.506,0:01:54.647
So think about the opposite approach,

0:01:54.647,0:01:58.178
being softer than[br]anything else around you.

0:01:58.178,0:02:03.136
Well, maybe you think that you're not[br]really able to do anything if you're soft,

0:02:03.136,0:02:04.150
probably.

0:02:04.150,0:02:07.312
Well, nature teaches us the opposite.

0:02:07.312,0:02:10.323
For example, at the bottom of the ocean[br]under thousands of pounds

0:02:10.323,0:02:11.611
of ?? pressure,

0:02:11.611,0:02:14.168
a completely soft animal

0:02:14.168,0:02:18.078
can move and interact with a much[br]stiffer object than him.

0:02:18.078,0:02:20.949
He works by carrying around[br]this coconut shell

0:02:20.949,0:02:23.357
thanks to the flexibility[br]of his tentacles,

0:02:23.357,0:02:26.465
which serve as both his feet and hands.

0:02:26.465,0:02:30.820
And apparently, an octopus[br]can also open a jar.

0:02:30.820,0:02:33.641
It's pretty impressive, right?

0:02:35.587,0:02:40.354
But clearly, this is not enabled[br]just by the brain of this animal,

0:02:40.354,0:02:42.614
but also by his body,

0:02:42.614,0:02:46.895
and it's a clear example,[br]maybe the clearest example,

0:02:46.895,0:02:48.608
of embodied intelligence,

0:02:48.608,0:02:51.994
which is a kind of intelligence[br]that all living organisms have.

0:02:51.994,0:02:53.309
We all have that.

0:02:53.309,0:02:57.326
Our body, its shape,[br]material and structure,

0:02:57.326,0:03:00.597
plays a fundamental role[br]during a physical task,

0:03:00.597,0:03:05.702
because we can conform[br]to our environment

0:03:05.702,0:03:08.660
so we can succeed in a large[br]variety of situations

0:03:08.660,0:03:11.614
without much planning[br]or calculations ahead.

0:03:11.614,0:03:15.445
So why don't we put[br]some of this embodied intelligence

0:03:15.445,0:03:16.649
into our robotic machines

0:03:16.649,0:03:18.559
to release them from relying[br]on excessive work

0:03:18.559,0:03:20.868
on computation and sensing?

0:03:20.868,0:03:23.918
Well, to do that we can follow[br]the strategy of nature,

0:03:23.918,0:03:26.740
because with evolution,[br]she's done a pretty good job

0:03:26.740,0:03:31.259
in designing machines[br]for environment interaction,

0:03:31.259,0:03:35.776
and it's easy to notice that nature[br]uses soft material frequently

0:03:35.776,0:03:38.094
and stiff material sparingly.

0:03:38.094,0:03:41.909
And this is what is done[br]in this new field or robotics

0:03:41.909,0:03:44.232
which is called soft robotics,

0:03:44.232,0:03:47.991
in which the main objective[br]is not to make super-precise machines

0:03:47.991,0:03:49.825
because we've already got them,

0:03:49.825,0:03:54.117
but to make robots able to face[br]unexpected situations in the real world,

0:03:54.117,0:03:56.562
so able to go out there.

0:03:56.562,0:03:59.994
And what makes a robot soft[br]is first of all his compliant body,

0:03:59.994,0:04:05.229
which is made of materials or structures[br]that can undergo very large deformations,

0:04:05.229,0:04:07.308
so no more rigid links,

0:04:07.308,0:04:11.366
and secondly to move them[br]we use what we call distributed actuation,

0:04:11.366,0:04:15.973
so we have to control continuously[br]the shape of this very deformable body,

0:04:15.973,0:04:19.294
which is the fact of having[br]a lot of links and joints,

0:04:19.294,0:04:22.063
but we don't have[br]any stiff structure at all.

0:04:22.063,0:04:25.305
So you can imagine that building[br]a soft robot is a very different process

0:04:25.305,0:04:29.203
than stiff robotics, where[br]you have links, gears, screws

0:04:29.203,0:04:31.458
that you must combine[br]in a very defined way.

0:04:31.458,0:04:35.006
In soft robots, you just build[br]your actuator from scratch

0:04:35.006,0:04:36.143
most of the time,

0:04:36.143,0:04:38.394
but you shape your flexible material

0:04:38.394,0:04:41.254
to the form that responds[br]to a certain input.

0:04:41.254,0:04:44.538
For example here, you can just[br]deform a structure

0:04:44.538,0:04:46.393
doing a fairly complex shape

0:04:46.393,0:04:49.533
if you think about doing the same[br]with rigid links and joints,

0:04:49.533,0:04:52.088
and here what you use is just one input,

0:04:52.088,0:04:54.069
such as air pressure.

0:04:54.069,0:04:57.582
Okay, but let's see[br]some cool examples of soft robots.

0:04:57.582,0:05:00.921
Here is a little cute guy[br]developed by Harvard University,

0:05:00.921,0:05:07.121
and he works thanks to waves[br]of pressure applied along its body,

0:05:07.121,0:05:10.568
and thanks to the flexibility he can[br]also sneak under a low bridge,

0:05:10.568,0:05:11.723
keep walking,

0:05:11.723,0:05:15.434
and then keep walking[br]a little bit different afterwards.

0:05:15.434,0:05:18.233
And it's a very preliminary prototype,

0:05:18.233,0:05:21.076
but they also built a more robust version

0:05:21.076,0:05:22.225
with power on board that can actually[br]be sent out in the world

0:05:22.225,0:05:26.971
and face real-world interactions

0:05:26.971,0:05:29.046
like a car passing it over it,

0:05:29.046,0:05:30.764
and keep working.

0:05:30.764,0:05:32.195
(Laughter)

0:05:32.195,0:05:33.477
It's cute.

0:05:33.477,0:05:34.983
(Laughter)

0:05:34.983,0:05:38.625
Or a robotic fish which swims[br]like a real fish does

0:05:38.625,0:05:42.486
in water simply because it has a soft tail[br]with distributed actuation

0:05:42.486,0:05:44.154
using still air pressure.

0:05:44.154,0:05:45.651
That was from MIT,

0:05:45.651,0:05:48.365
and of course we have a robotic octopus.

0:05:48.365,0:05:50.735
This was actually one of[br]the first projects developed

0:05:50.735,0:05:52.618
in this new field of soft robots.

0:05:52.618,0:05:54.733
Here you see the artificial tentacle,

0:05:54.733,0:05:58.343
but they actually built and entire machine

0:05:58.343,0:06:01.999
with several tentacles they could[br]just throw in the water,

0:06:01.999,0:06:05.319
and you see that it can kind of go around[br]and do submarine exploration

0:06:05.319,0:06:09.577
in a different way[br]than rigid robots would do.

0:06:09.577,0:06:13.194
But this is very important for delicate[br]environments such as coral reefs.

0:06:13.194,0:06:15.188
Let's go back to the ground.

0:06:15.188,0:06:17.687
Here you see the view

0:06:17.687,0:06:20.000
from a growing robot developed[br]by my colleagues in Stanford.

0:06:20.000,0:06:21.970
You see the camera fixed on top.

0:06:21.970,0:06:25.070
And this robot is particular[br]because using air pressure

0:06:25.070,0:06:27.134
it grows from the tip[br]while the rest of the body

0:06:27.134,0:06:29.627
stays in firm contact[br]with the environment.

0:06:29.627,0:06:32.368
And this is inspired[br]by plants, not animals,

0:06:32.368,0:06:35.597
which grows via the material[br]in a similar manner

0:06:35.597,0:06:39.172
so it can face a pretty large[br]variety of situations.

0:06:39.172,0:06:41.009
But I'm a biomedical engineer,

0:06:41.009,0:06:43.459
and perhaps the application[br]I like the most

0:06:43.459,0:06:45.016
it's in the medical field,

0:06:45.016,0:06:49.929
and it's very difficult to imagine[br]a closer interaction with the human body

0:06:49.929,0:06:51.744
than actually going inside the body,

0:06:51.744,0:06:55.294
for example to inform[br]a minimally invasive procedure.

0:06:55.294,0:06:58.680
And here, robots can be[br]very helpful with the surgeon

0:06:58.680,0:07:00.600
because they must enter the body

0:07:00.600,0:07:03.376
using small holes[br]and straight instruments,

0:07:03.376,0:07:06.542
and these instruments must interact[br]with very delicate structures

0:07:06.542,0:07:08.717
in a very uncertain environment,

0:07:08.717,0:07:10.678
and this must be done safely.

0:07:10.678,0:07:13.022
Also bringing the camera inside the body,

0:07:13.022,0:07:17.126
so bringing the eyes of the surgeon[br]inside the subject, I feel,

0:07:17.126,0:07:18.791
can be very challenging[br]if you use a rigid stick,

0:07:18.791,0:07:20.422
like a classic endoscope.

0:07:20.422,0:07:23.370
With my previous research group in Europe,

0:07:23.370,0:07:26.175
we developed this[br]self-camera robot for surgery,

0:07:26.175,0:07:29.500
which is very different[br]from a classic endoscope

0:07:29.500,0:07:34.116
which can move thanks[br]to the flexibility of the module

0:07:34.116,0:07:35.710
that can bend and elongate,

0:07:35.710,0:07:37.955
bend in every direction[br]and also elongate.

0:07:37.955,0:07:41.494
And this was actually used by surgeons[br]to see what they were doing

0:07:41.494,0:07:43.443
with other instruments[br]from different points of view

0:07:43.443,0:07:47.479
without caring that much[br]about what was touched around.

0:07:47.479,0:07:51.245
And here you see the soft robot in action,

0:07:51.245,0:07:54.576
and it just goes inside.

0:07:54.576,0:07:57.461
This is a body simulator,[br]not a real human body.

0:07:57.461,0:07:58.597
It goes around.

0:07:58.597,0:08:01.076
You have a light, because usually[br]you don't have too many lights

0:08:01.076,0:08:02.178
inside your body.

0:08:02.178,0:08:03.423
(Laughter)

0:08:03.423,0:08:04.588
We hope.

0:08:04.588,0:08:07.970
(Laughter)

0:08:07.970,0:08:12.222
But sometimes a surgical procedure[br]can even be done using a single needle,

0:08:12.222,0:08:15.716
and in Stanford now, we are working[br]on a very flexible needle,

0:08:15.716,0:08:19.124
kind of a very tiny soft robot

0:08:19.124,0:08:23.709
which is mechanically designed[br]to use the interaction with the tissue

0:08:23.709,0:08:24.837
and steer around inside a solid organ.

0:08:24.837,0:08:29.068
This makes it possible to reach[br]many different targets such as tumors

0:08:29.068,0:08:32.264
deep inside a solid organ

0:08:32.264,0:08:32.987
by using one single insertion point,

0:08:32.987,0:08:39.534
and you can even steer around[br]the structure that you want to avoid

0:08:39.534,0:08:39.983
on the way to the target.

0:08:39.983,0:08:42.946
So clearly, this is a pretty[br]exciting time for robotics.

0:08:42.946,0:08:46.083
We have robots that have to deal[br]with soft structure,

0:08:46.083,0:08:50.238
so this poses new[br]and very challenging questions

0:08:50.238,0:08:51.204
for the robotics community,

0:08:51.204,0:08:52.968
and indeed we are just starting[br]to learn how to control,

0:08:52.968,0:08:55.935
how to put sensors[br]on these very flexible structures.

0:08:55.935,0:08:58.619
But of course, we are not even close[br]to what nature figured out

0:08:58.619,0:09:01.002
in millions of years of evolution.

0:09:01.002,0:09:03.303
But one thing I know for sure:

0:09:03.303,0:09:05.842
that robots will be softer and safer,

0:09:05.842,0:09:08.375
and they will be out there helping people.

0:09:08.375,0:09:09.191
Yeah.

0:09:09.191,0:09:11.257
Thank you.

0:09:11.257,0:09:14.516
(Applause)