0:00:00.809,0:00:03.983
So the first robot[br]to talk about is called STriDER.

0:00:04.007,0:00:07.476
It stands for Self-excited[br]Tripedal Dynamic Experimental Robot.

0:00:07.500,0:00:11.858
It's a robot that has three legs,[br]which is inspired by nature.

0:00:12.500,0:00:15.911
But have you seen anything in nature,[br]an animal that has three legs?

0:00:15.935,0:00:19.805
Probably not. So why do I call this[br]a biologically inspired robot?

0:00:19.829,0:00:20.991
How would it work?

0:00:21.015,0:00:23.177
But before that,[br]let's look at pop culture.

0:00:23.201,0:00:26.669
So, you know H.G. Wells's[br]"War of the Worlds," novel and movie.

0:00:26.693,0:00:30.014
And what you see over here[br]is a very popular video game,

0:00:30.038,0:00:33.738
and in this fiction, they describe[br]these alien creatures and robots

0:00:33.762,0:00:36.024
that have three legs that terrorize Earth.

0:00:36.048,0:00:39.583
But my robot, STriDER,[br]does not move like this.

0:00:39.607,0:00:42.658
This is an actual dynamic[br]simulation animation.

0:00:42.682,0:00:44.739
I'm going to show you how the robot works.

0:00:44.763,0:00:47.155
It flips its body 180 degrees

0:00:47.179,0:00:49.237
and it swings its leg between the two legs

0:00:49.261,0:00:50.420
and catches the fall.

0:00:50.444,0:00:51.606
So that's how it walks.

0:00:51.630,0:00:54.342
But when you look at us[br]human beings, bipedal walking,

0:00:54.366,0:00:55.524
what you're doing is,

0:00:55.548,0:00:59.635
you're not really using muscle[br]to lift your leg and walk like a robot.

0:00:59.659,0:01:02.996
What you're doing is,[br]you swing your leg and catch the fall,

0:01:03.020,0:01:05.476
stand up again, swing your leg[br]and catch the fall.

0:01:05.500,0:01:08.667
You're using your built-in dynamics,[br]the physics of your body,

0:01:08.691,0:01:10.442
just like a pendulum.

0:01:10.466,0:01:13.824
We call that the concept[br]of passive dynamic locomotion.

0:01:13.848,0:01:15.986
What you're doing is, when you stand up,

0:01:16.010,0:01:18.091
potential energy to kinetic energy,

0:01:18.115,0:01:20.092
potential energy to kinetic energy.

0:01:20.116,0:01:22.093
It's a constantly falling process.

0:01:22.117,0:01:25.166
So even though there is nothing[br]in nature that looks like this,

0:01:25.190,0:01:29.319
really, we're inspired by biology[br]and applying the principles of walking

0:01:29.343,0:01:30.500
to this robot.

0:01:30.524,0:01:32.501
Thus, it's a biologically inspired robot.

0:01:32.525,0:01:34.938
What you see here,[br]this is what we want to do next.

0:01:34.962,0:01:38.476
We want to fold up the legs[br]and shoot it up for long-range motion.

0:01:38.500,0:01:41.294
And it deploys legs --[br]it looks almost like "Star Wars" --

0:01:41.318,0:01:44.182
so when it lands, it absorbs[br]the shock and starts walking.

0:01:44.206,0:01:47.411
What you see over here, this yellow thing,[br]this is not a death ray.

0:01:47.435,0:01:48.447
(Laughter)

0:01:48.471,0:01:49.630
This is just to show you

0:01:49.654,0:01:52.341
that if you have cameras[br]or different types of sensors,

0:01:52.365,0:01:53.882
because it's 1.8 meters tall,

0:01:53.906,0:01:56.992
you can see over obstacles like bushes[br]and those kinds of things.

0:01:57.016,0:01:58.363
So we have two prototypes.

0:01:58.387,0:02:01.330
The first version,[br]in the back, that's STriDER I.

0:02:01.354,0:02:03.521
The one in front,[br]the smaller, is STriDER II.

0:02:03.545,0:02:07.053
The problem we had with STriDER I is,[br]it was just too heavy in the body.

0:02:07.077,0:02:09.661
We had so many motors aligning the joints

0:02:09.685,0:02:10.939
and those kinds of things.

0:02:10.963,0:02:14.163
So we decided to synthesize[br]a mechanical mechanism

0:02:14.187,0:02:17.256
so we could get rid of all the motors,[br]and with a single motor,

0:02:17.280,0:02:18.957
we can coordinate all the motions.

0:02:18.981,0:02:22.358
It's a mechanical solution to a problem,[br]instead of using mechatronics.

0:02:22.382,0:02:25.794
So with this, now the top body[br]is lighted up; it's walking in our lab.

0:02:25.818,0:02:27.877
This was the very first successful step.

0:02:27.901,0:02:30.569
It's still not perfected,[br]its coffee falls down,

0:02:30.593,0:02:32.575
so we still have a lot of work to do.

0:02:33.425,0:02:36.121
The second robot I want[br]to talk about is called IMPASS.

0:02:36.145,0:02:40.604
It stands for Intelligent Mobility[br]Platform with Actuated Spoke System.

0:02:40.628,0:02:43.223
It's a wheel-leg hybrid robot.

0:02:43.247,0:02:47.104
So think of a rimless wheel[br]or a spoke wheel,

0:02:47.128,0:02:50.105
but the spokes individually[br]move in and out of the hub;

0:02:50.129,0:02:52.216
so, it's a wheel-leg hybrid.

0:02:52.240,0:02:54.482
We're literally reinventing[br]the wheel here.

0:02:54.506,0:02:56.969
Let me demonstrate how it works.

0:02:56.993,0:03:00.842
So in this video we're using an approach[br]called the reactive approach.

0:03:00.866,0:03:03.843
Just simply using[br]the tactile sensors on the feet,

0:03:03.867,0:03:06.677
it's trying to walk[br]over a changing terrain,

0:03:06.701,0:03:09.401
a soft terrain where it pushes[br]down and changes.

0:03:09.425,0:03:11.622
And just by the tactile information,

0:03:11.646,0:03:14.401
it successfully crosses[br]over these types of terrains.

0:03:14.425,0:03:18.230
But, when it encounters[br]a very extreme terrain --

0:03:18.254,0:03:22.167
in this case, this obstacle[br]is more than three times the height

0:03:22.191,0:03:23.361
of the robot --

0:03:23.385,0:03:25.259
then it switches to a deliberate mode,

0:03:25.283,0:03:28.097
where it uses a laser range finder[br]and camera systems

0:03:28.121,0:03:29.947
to identify the obstacle and the size.

0:03:29.971,0:03:32.955
And it carefully plans[br]the motion of the spokes

0:03:32.979,0:03:36.688
and coordinates it so it can show[br]this very impressive mobility.

0:03:36.712,0:03:39.353
You probably haven't seen[br]anything like this out there.

0:03:39.377,0:03:43.047
This is a very high-mobility robot[br]that we developed called IMPASS.

0:03:44.613,0:03:46.401
Ah, isn't that cool?

0:03:46.425,0:03:49.719
When you drive your car,

0:03:49.743,0:03:53.665
when you steer your car, you use[br]a method called Ackermann steering.

0:03:53.689,0:03:55.540
The front wheels rotate like this.

0:03:55.564,0:03:58.079
For most small-wheeled robots,

0:03:58.103,0:04:00.352
they use a method[br]called differential steering

0:04:00.376,0:04:03.233
where the left and right wheel[br]turn the opposite direction.

0:04:03.257,0:04:06.155
For IMPASS, we can do many,[br]many different types of motion.

0:04:06.179,0:04:07.448
For example, in this case,

0:04:07.472,0:04:09.872
even though the left and right[br]wheels are connected

0:04:09.896,0:04:12.662
with a single axle rotating[br]at the same angle of velocity,

0:04:12.686,0:04:15.820
we simply change the length[br]of the spoke, it affects the diameter,

0:04:15.844,0:04:17.931
then can turn to the left[br]and to the right.

0:04:17.955,0:04:21.308
These are just some examples[br]of the neat things we can do with IMPASS.

0:04:21.332,0:04:23.106
This robot is called CLIMBeR:

0:04:23.130,0:04:26.530
Cable-suspended Limbed Intelligent[br]Matching Behavior Robot.

0:04:26.554,0:04:29.681
I've been talking[br]to a lot of NASA JPL scientists --

0:04:29.705,0:04:31.855
at JPL, they are famous[br]for the Mars rovers --

0:04:31.879,0:04:34.250
and the scientists,[br]geologists always tell me

0:04:34.274,0:04:37.401
that the real interesting science,[br]the science-rich sites,

0:04:37.425,0:04:38.936
are always at the cliffs.

0:04:38.960,0:04:40.910
But the current rovers cannot get there.

0:04:40.934,0:04:43.417
So, inspired by that,[br]we wanted to build a robot

0:04:43.441,0:04:46.558
that can climb[br]a structured cliff environment.

0:04:46.582,0:04:47.806
So this is CLIMBeR.

0:04:47.830,0:04:49.341
It has three legs.

0:04:49.365,0:04:52.948
It's probably difficult to see, but it has[br]a winch and a cable at the top.

0:04:52.972,0:04:55.553
It tries to figure out[br]the best place to put its foot.

0:04:55.577,0:04:57.212
And then once it figures that out,

0:04:57.236,0:05:00.213
in real time, it calculates[br]the force distribution:

0:05:00.237,0:05:02.760
how much force it needs[br]to exert to the surface

0:05:02.784,0:05:04.761
so it doesn't tip and doesn't slip.

0:05:04.785,0:05:06.926
Once it stabilizes that, it lifts a foot,

0:05:06.950,0:05:10.253
and then with the winch,[br]it can climb up these kinds of cliffs.

0:05:10.785,0:05:13.401
Also for search and rescue[br]applications as well.

0:05:13.425,0:05:15.844
Five years ago,[br]I actually worked at NASA JPL

0:05:15.868,0:05:17.718
during the summer as a faculty fellow.

0:05:17.742,0:05:21.192
And they already had[br]a six-legged robot called LEMUR.

0:05:21.216,0:05:22.856
So this is actually based on that.

0:05:22.880,0:05:24.171
This robot is called MARS:

0:05:24.195,0:05:25.812
Multi-Appendage Robotic System.

0:05:25.836,0:05:27.074
It's a hexapod robot.

0:05:27.098,0:05:29.075
We developed our adaptive gait planner.

0:05:29.099,0:05:31.624
We actually have a very interesting[br]payload on there.

0:05:31.648,0:05:33.095
The students like to have fun.

0:05:33.119,0:05:36.214
And here you can see that it's walking[br]over unstructured terrain.

0:05:36.238,0:05:37.355
(Motor sound)

0:05:37.379,0:05:40.176
It's trying to walk[br]on the coastal terrain, a sandy area,

0:05:40.200,0:05:44.848
but depending on the moisture content[br]or the grain size of the sand,

0:05:44.872,0:05:48.716
the foot's soil sinkage model changes,[br]so it tries to adapt its gait

0:05:48.740,0:05:50.980
to successfully cross[br]over these kind of things.

0:05:51.004,0:05:52.511
It also does some fun stuff.

0:05:52.535,0:05:56.020
As you can imagine,[br]we get so many visitors visiting our lab.

0:05:56.044,0:05:58.776
So when the visitors come,[br]MARS walks up to the computer,

0:05:58.800,0:06:00.848
starts typing, "Hello, my name is MARS.

0:06:00.872,0:06:02.401
Welcome to RoMeLa,

0:06:02.425,0:06:05.038
the Robotics Mechanisms[br]Laboratory at Virginia Tech."

0:06:05.062,0:06:06.532
(Laughter)

0:06:06.556,0:06:08.681
This robot is an amoeba robot.

0:06:08.705,0:06:11.593
Now, we don't have enough time[br]to go into technical details,

0:06:11.617,0:06:13.674
I'll just show you[br]some of the experiments.

0:06:13.698,0:06:16.210
These are some of the early[br]feasibility experiments.

0:06:16.234,0:06:19.319
We store potential energy[br]to the elastic skin to make it move,

0:06:19.343,0:06:23.146
or use active tension cords[br]to make it move forward and backward.

0:06:23.170,0:06:24.327
It's called ChIMERA.

0:06:24.351,0:06:27.363
We also have been working[br]with some scientists and engineers

0:06:27.387,0:06:28.538
from UPenn

0:06:28.562,0:06:32.395
to come up with a chemically actuated[br]version of this amoeba robot.

0:06:32.419,0:06:34.302
We do something to something,

0:06:34.326,0:06:36.994
and just like magic, it moves.

0:06:37.626,0:06:39.208
"The Blob."

0:06:39.933,0:06:41.710
This robot is a very recent project.

0:06:41.734,0:06:42.885
It's called RAPHaEL:

0:06:42.909,0:06:45.596
Robotic Air-Powered Hand[br]with Elastic Ligaments.

0:06:45.620,0:06:48.619
There are a lot of really neat,[br]very good robotic hands

0:06:48.643,0:06:49.806
out there on the market.

0:06:49.830,0:06:52.010
The problem is,[br]they're just too expensive --

0:06:52.034,0:06:53.439
tens of thousands of dollars.

0:06:53.463,0:06:56.484
So for prosthesis applications[br]it's probably not too practical,

0:06:56.508,0:06:57.853
because it's not affordable.

0:06:57.877,0:07:01.351
We wanted to tackle this problem[br]in a very different direction.

0:07:01.375,0:07:04.424
Instead of using electrical motors,[br]electromechanical actuators,

0:07:04.448,0:07:06.204
we're using compressed air.

0:07:06.228,0:07:09.675
We developed these novel actuators[br]for the joints, so it's compliant.

0:07:09.699,0:07:11.401
You can actually change the force,

0:07:11.425,0:07:13.251
simply just changing the air pressure.

0:07:13.275,0:07:15.465
And it can actually crush[br]an empty soda can.

0:07:15.489,0:07:18.539
It can pick up very delicate[br]objects like a raw egg,

0:07:18.563,0:07:20.151
or in this case, a lightbulb.

0:07:20.790,0:07:24.459
The best part: it took only 200 dollars[br]to make the first prototype.

0:07:25.906,0:07:28.654
This robot is actually[br]a family of snake robots

0:07:28.678,0:07:30.046
that we call HyDRAS,

0:07:30.070,0:07:32.781
Hyper Degrees-of-freedom Robotic[br]Articulated Serpentine.

0:07:32.805,0:07:34.976
This is a robot that can climb structures.

0:07:35.000,0:07:37.083
This is a HyDRAS's arm.

0:07:37.107,0:07:39.116
It's a 12-degrees-of-freedom robotic arm.

0:07:39.140,0:07:41.488
But the cool part is the user interface.

0:07:41.512,0:07:44.255
The cable over there,[br]that's an optical fiber.

0:07:44.279,0:07:46.715
This student, it's probably[br]her first time using it,

0:07:46.739,0:07:49.096
but she can articulate it[br]in many different ways.

0:07:49.120,0:07:52.583
So, for example, in Iraq, the war zone,[br]there are roadside bombs.

0:07:52.607,0:07:56.618
Currently, you send these remotely[br]controlled vehicles that are armed.

0:07:56.642,0:08:00.181
It takes really a lot of time[br]and it's expensive to train the operator

0:08:00.205,0:08:01.943
to operate this complex arm.

0:08:01.967,0:08:03.934
In this case, it's very intuitive;

0:08:03.958,0:08:06.226
this student, probably[br]his first time using it,

0:08:06.250,0:08:08.228
is doing very complex manipulation tasks,

0:08:08.252,0:08:11.613
picking up objects and doing[br]manipulation, just like that.

0:08:11.637,0:08:12.842
Very intuitive.

0:08:14.765,0:08:17.362
Now, this robot[br]is currently our star robot.

0:08:17.386,0:08:20.371
We actually have a fan club[br]for the robot, DARwIn:

0:08:20.395,0:08:23.296
Dynamic Anthropomorphic[br]Robot with Intelligence.

0:08:23.320,0:08:27.203
As you know, we're very interested[br]in human walking,

0:08:27.227,0:08:29.419
so we decided to build[br]a small humanoid robot.

0:08:29.443,0:08:31.205
This was in 2004; at that time,

0:08:31.229,0:08:33.499
this was something really,[br]really revolutionary.

0:08:33.523,0:08:35.324
This was more of a feasibility study:

0:08:35.348,0:08:37.999
What kind of motors should we use?[br]Is it even possible?

0:08:38.023,0:08:39.757
What kinds of controls should we do?

0:08:39.781,0:08:43.301
This does not have any sensors,[br]so it's an open-loop control.

0:08:43.325,0:08:46.095
For those who probably know,[br]if you don't have any sensors

0:08:46.119,0:08:48.692
and there's any disturbances,[br]you know what happens.

0:08:48.716,0:08:50.829
(Laughter)

0:08:50.853,0:08:56.077
Based on that success, the following year[br]we did the proper mechanical design,

0:08:56.101,0:08:57.325
starting from kinematics.

0:08:57.349,0:09:00.259
And thus, DARwIn I was born in 2005.

0:09:00.283,0:09:02.594
It stands up, it walks -- very impressive.

0:09:02.618,0:09:05.975
However, still, as you can see,[br]it has a cord, an umbilical cord.

0:09:05.999,0:09:08.171
So we're still using[br]an external power source

0:09:08.195,0:09:09.794
and external computation.

0:09:10.742,0:09:14.053
So in 2006, now it's really[br]time to have fun.

0:09:14.077,0:09:15.536
Let's give it intelligence.

0:09:15.560,0:09:17.681
We give it all the computing[br]power it needs:

0:09:17.705,0:09:20.235
a 1.5 gigahertz Pentium M chip,[br]two FireWire cameras,

0:09:20.259,0:09:23.105
rate gyros, accelerometers,[br]four forced sensors on the foot,

0:09:23.129,0:09:24.479
lithium polymer batteries --

0:09:24.503,0:09:27.653
and now DARwIn II[br]is completely autonomous.

0:09:28.146,0:09:30.837
It is not remote controlled.[br]There's no tethers.

0:09:30.861,0:09:34.594
It looks around, searches for the ball ...[br]looks around, searches for the ball,

0:09:34.618,0:09:39.547
and it tries to play a game of soccer[br]autonomously -- artificial intelligence.

0:09:39.571,0:09:40.725
Let's see how it does.

0:09:40.749,0:09:42.696
This was our very first trial, and ...

0:09:42.720,0:09:47.087
(Video) Spectators: Goal!

0:09:48.238,0:09:51.188
Dennis Hong: There is actually[br]a competition called RoboCup.

0:09:51.212,0:09:53.799
I don't know how many of you[br]have heard about RoboCup.

0:09:53.823,0:09:58.197
It's an international autonomous[br]robot soccer competition.

0:09:58.221,0:10:00.969
And the actual goal of RoboCup is,

0:10:00.993,0:10:03.186
by the year 2050,

0:10:03.210,0:10:06.941
we want to have full-size,[br]autonomous humanoid robots

0:10:06.965,0:10:10.039
play soccer against the human[br]World Cup champions

0:10:10.063,0:10:11.216
and win.

0:10:11.240,0:10:12.272
(Laughter)

0:10:12.296,0:10:13.514
It's a true, actual goal.

0:10:13.538,0:10:17.435
It's a very ambitious goal,[br]but we truly believe we can do it.

0:10:17.459,0:10:19.193
This is last year in China.

0:10:19.217,0:10:22.230
We were the very first team[br]in the United States that qualified

0:10:22.254,0:10:24.054
in the humanoid RoboCup competition.

0:10:24.078,0:10:26.258
This is this year in Austria.

0:10:26.282,0:10:28.849
You're going to see the action[br]is three against three,

0:10:28.873,0:10:30.064
completely autonomous.

0:10:30.088,0:10:31.179
(Video) (Crowd groans)

0:10:31.203,0:10:32.533
DH: There you go. Yes!

0:10:33.331,0:10:37.473
The robots track and they team-play[br]amongst themselves.

0:10:37.934,0:10:39.085
It's very impressive.

0:10:39.109,0:10:40.585
It's really a research event,

0:10:40.609,0:10:44.849
packaged in a more exciting[br]competition event.

0:10:44.873,0:10:48.401
What you see here is the beautiful[br]Louis Vuitton Cup trophy.

0:10:48.425,0:10:49.944
This is for the best humanoid.

0:10:49.968,0:10:53.637
We'd like to bring this, for the first[br]time, to the United States next year,

0:10:53.661,0:10:54.816
so wish us luck.

0:10:54.840,0:10:55.876
(Applause)

0:10:55.900,0:10:57.052
Thank you.

0:10:57.076,0:10:59.157
(Applause)

0:10:59.181,0:11:01.249
DARwIn also has a lot of other talents.

0:11:01.273,0:11:04.989
Last year, it actually conducted[br]the Roanoke Symphony Orchestra

0:11:05.013,0:11:07.401
for the holiday concert.

0:11:07.425,0:11:10.401
This is the next generation[br]robot, DARwIn IV,

0:11:10.425,0:11:13.401
much smarter, faster, stronger.

0:11:13.425,0:11:15.402
And it's trying to show off its ability:

0:11:15.426,0:11:17.223
"I'm macho, I'm strong."

0:11:17.247,0:11:18.691
(Laughter)

0:11:18.715,0:11:22.940
"I can also do some Jackie Chan-motion,[br]martial art movements."

0:11:22.964,0:11:24.976
(Laughter)

0:11:26.425,0:11:28.360
And it walks away. So this is DARwIn IV.

0:11:28.384,0:11:30.519
Again, you'll be able[br]to see it in the lobby.

0:11:30.543,0:11:33.735
We truly believe this will be[br]the very first running humanoid robot

0:11:33.759,0:11:34.910
in the United States.

0:11:34.934,0:11:36.088
So stay tuned.

0:11:36.112,0:11:39.113
All right. So I showed you[br]some of our exciting robots at work.

0:11:39.137,0:11:41.325
So, what is the secret of our success?

0:11:41.349,0:11:43.166
Where do we come up with these ideas?

0:11:43.190,0:11:45.083
How do we develop these kinds of ideas?

0:11:45.107,0:11:46.897
We have a fully autonomous vehicle

0:11:46.921,0:11:48.797
that can drive into urban environments.

0:11:48.821,0:11:51.720
We won a half a million dollars[br]in the DARPA Urban Challenge.

0:11:51.744,0:11:55.401
We also have the world's very first[br]vehicle that can be driven by the blind.

0:11:55.425,0:11:57.949
We call it the Blind Driver[br]Challenge, very exciting.

0:11:57.973,0:12:01.145
And many, many other robotics[br]projects I want to talk about.

0:12:01.169,0:12:03.806
These are just the awards[br]that we won in 2007 fall

0:12:03.830,0:12:06.401
from robotics competitions[br]and those kinds of things.

0:12:06.425,0:12:08.318
So really, we have five secrets.

0:12:08.342,0:12:10.778
First is: Where do we get inspiration?

0:12:10.802,0:12:12.881
Where do we get this spark of imagination?

0:12:12.905,0:12:14.926
This is a true story, my personal story.

0:12:14.950,0:12:17.727
At night, when I go to bed,[br]at three, four in the morning,

0:12:17.751,0:12:20.640
I lie down, close my eyes,[br]and I see these lines and circles

0:12:20.664,0:12:22.458
and different shapes floating around.

0:12:22.482,0:12:25.245
And they assemble, and they form[br]these kinds of mechanisms.

0:12:25.269,0:12:26.877
And I think, "Ah, this is cool."

0:12:26.901,0:12:29.778
So right next to my bed[br]I keep a notebook, a journal,

0:12:29.802,0:12:32.158
with a special pen[br]that has an LED light on it,

0:12:32.182,0:12:35.219
because I don't want to turn on the light[br]and wake up my wife.

0:12:35.243,0:12:38.512
So I see this, scribble everything down,[br]draw things, and go to bed.

0:12:38.536,0:12:40.926
Every day in the morning,[br]the first thing I do,

0:12:40.950,0:12:43.557
before my first cup of coffee,[br]before I brush my teeth,

0:12:43.581,0:12:44.743
I open my notebook.

0:12:44.767,0:12:47.430
Many times it's empty;[br]sometimes I have something there.

0:12:47.454,0:12:49.520
If something's there, sometimes it's junk.

0:12:49.544,0:12:51.901
But most of the time,[br]I can't read my handwriting.

0:12:51.925,0:12:54.249
Four in the morning --[br]what do you expect, right?

0:12:54.273,0:12:56.177
So I need to decipher what I wrote.

0:12:56.201,0:12:59.401
But sometimes I see[br]this ingenious idea in there,

0:12:59.425,0:13:00.910
and I have this eureka moment.

0:13:00.934,0:13:03.450
I directly run to my home office,[br]sit at my computer,

0:13:03.474,0:13:05.451
I type in the ideas, I sketch things out

0:13:05.475,0:13:07.366
and I keep a database of ideas.

0:13:08.028,0:13:10.389
So when we have these calls for proposals,

0:13:10.413,0:13:13.983
I try to find a match[br]between my potential ideas

0:13:14.007,0:13:15.167
and the problem.

0:13:15.191,0:13:17.508
If there's a match,[br]we write a research proposal,

0:13:17.532,0:13:18.879
get the research funding in,

0:13:18.903,0:13:21.063
and that's how we start[br]our research programs.

0:13:21.087,0:13:23.580
But just a spark of imagination[br]is not good enough.

0:13:23.604,0:13:25.484
How do we develop these kinds of ideas?

0:13:25.508,0:13:28.288
At our lab RoMeLa, the Robotics[br]and Mechanisms Laboratory,

0:13:28.312,0:13:30.924
we have these fantastic[br]brainstorming sessions.

0:13:30.948,0:13:35.283
So we gather around, we discuss problems[br]and solutions and talk about it.

0:13:35.307,0:13:38.284
But before we start,[br]we set this golden rule.

0:13:38.308,0:13:39.957
The rule is:

0:13:39.981,0:13:42.958
nobody criticizes anybody's ideas.

0:13:42.982,0:13:44.987
Nobody criticizes any opinion.

0:13:45.425,0:13:48.972
This is important, because many times,[br]students fear or feel uncomfortable

0:13:48.996,0:13:52.265
about how others might think[br]about their opinions and thoughts.

0:13:52.289,0:13:56.079
So once you do this, it is amazing[br]how the students open up.

0:13:56.103,0:13:59.401
They have these wacky, cool,[br]crazy, brilliant ideas,

0:13:59.425,0:14:02.854
and the whole room is just electrified[br]with creative energy.

0:14:02.878,0:14:05.151
And this is how we develop our ideas.

0:14:05.738,0:14:07.295
Well, we're running out of time.

0:14:07.319,0:14:09.221
One more thing I want to talk about is,

0:14:09.245,0:14:12.401
you know, just a spark of idea[br]and development is not good enough.

0:14:12.425,0:14:17.136
There was a great TED moment --[br]I think it was Sir Ken Robinson, was it?

0:14:17.160,0:14:21.598
He gave a talk about how education[br]and school kill creativity.

0:14:21.622,0:14:24.401
Well, actually,[br]there's two sides to the story.

0:14:24.425,0:14:29.638
So there is only so much one can do[br]with just ingenious ideas

0:14:29.662,0:14:32.678
and creativity[br]and good engineering intuition.

0:14:32.702,0:14:34.496
If you want to go beyond a tinkering,

0:14:34.520,0:14:36.670
if you want to go[br]beyond a hobby of robotics

0:14:36.694,0:14:40.135
and really tackle[br]the grand challenges of robotics

0:14:40.159,0:14:41.401
through rigorous research,

0:14:41.425,0:14:42.580
we need more than that.

0:14:42.604,0:14:44.618
This is where school comes in.

0:14:44.642,0:14:47.136
Batman, fighting against the bad guys,

0:14:47.160,0:14:49.553
he has his utility belt,[br]he has his grappling hook,

0:14:49.577,0:14:51.401
he has all different kinds of gadgets.

0:14:51.425,0:14:53.838
For us roboticists,[br]engineers and scientists,

0:14:53.862,0:14:58.019
these tools are the courses[br]and classes you take in class.

0:14:58.043,0:15:00.045
Math, differential equations.

0:15:00.069,0:15:02.809
I have linear algebra, science, physics --

0:15:02.833,0:15:05.801
even, nowadays, chemistry[br]and biology, as you've seen.

0:15:05.825,0:15:07.583
These are all the tools we need.

0:15:07.607,0:15:09.483
So the more tools you have, for Batman,

0:15:09.507,0:15:11.484
more effective at fighting the bad guys,

0:15:11.508,0:15:14.414
for us, more tools to attack[br]these kinds of big problems.

0:15:15.582,0:15:17.447
So education is very important.

0:15:18.568,0:15:21.165
Also -- it's not only about that.

0:15:21.189,0:15:23.257
You also have to work really, really hard.

0:15:23.281,0:15:24.734
So I always tell my students,

0:15:24.758,0:15:26.809
"Work smart, then work hard."

0:15:26.833,0:15:29.557
This picture in the back --[br]this is three in the morning.

0:15:29.581,0:15:31.910
I guarantee if you come[br]to our lab at 3, 4am,

0:15:31.934,0:15:33.490
we have students working there,

0:15:33.514,0:15:36.723
not because I tell them to,[br]but because we are having too much fun.

0:15:36.747,0:15:38.401
Which leads to the last topic:

0:15:38.425,0:15:40.294
do not forget to have fun.

0:15:40.318,0:15:43.531
That's really the secret of our success,[br]we're having too much fun.

0:15:43.555,0:15:47.021
I truly believe that highest productivity[br]comes when you're having fun,

0:15:47.045,0:15:48.401
and that's what we're doing.

0:15:48.425,0:15:49.579
And there you go.

0:15:49.603,0:15:50.770
Thank you so much.

0:15:50.794,0:15:54.896
(Applause)