WEBVTT

00:00:17.706 --> 00:00:19.267
Thanks for having me.

00:00:19.267 --> 00:00:23.231
We have too many really exciting
robotics works that I want to show you

00:00:23.231 --> 00:00:25.036
but we only have 18 minutes,

00:00:25.036 --> 00:00:28.391
so I really had a hard time
trying to cut down the slides.

00:00:28.391 --> 00:00:30.576
But let's see how it goes,
we have 18 minutes

00:00:30.576 --> 00:00:34.507
and an apology in advance,
I'm probably going to speak really fast.

00:00:34.507 --> 00:00:37.951
So, the first robot I'll talk about
is called STriDER.

00:00:37.951 --> 00:00:41.467
It stands for Self-excited
Tripedal Dynamic Experimental Robot.

00:00:41.467 --> 00:00:43.126
It's a robot that has three legs,

00:00:43.126 --> 00:00:45.856
which is inspired by nature.

00:00:45.856 --> 00:00:48.031
But have you seen anything in nature,

00:00:48.031 --> 00:00:49.871
an animal that has three legs?

00:00:49.871 --> 00:00:51.009
Probably not.

00:00:51.009 --> 00:00:53.554
So, why do I call this
a biologically inspired robot?

00:00:53.554 --> 00:00:54.755
How would it work?

00:00:54.755 --> 00:00:57.060
But before that,
let's look at pop culture.

00:00:57.060 --> 00:01:00.770
So, you know H.G. Wells'
"War of the Worlds," novel and movie.

00:01:00.770 --> 00:01:02.098
And what you see over here

00:01:02.098 --> 00:01:04.157
is a very popular video game,

00:01:04.157 --> 00:01:07.183
and in this fiction they describe
these alien creatures

00:01:07.183 --> 00:01:09.993
and robots that have three legs
that terrorize Earth.

00:01:09.993 --> 00:01:13.842
But my robot, STriDER,
does not move like this.

00:01:13.842 --> 00:01:15.252
So, how does it work?

00:01:15.252 --> 00:01:18.000
So, this is an actual
dynamic simulation animation.

00:01:18.000 --> 00:01:20.333
I'm just going to show you
how the robot works.

00:01:20.333 --> 00:01:22.139
So when I go to robotics conferences,

00:01:22.139 --> 00:01:24.167
I show this video to some of my colleagues

00:01:24.167 --> 00:01:26.919
and everybody goes, wow, this is cool.

00:01:26.919 --> 00:01:29.496
So when I click this,
it's going to show an animation,

00:01:29.496 --> 00:01:32.144
so everybody say "Ooh" and "Aah".

00:01:33.445 --> 00:01:35.052
Ooh.

00:01:36.575 --> 00:01:38.819
Aah. Isn't that cool?

00:01:40.165 --> 00:01:42.308
It flips its body 180 degrees

00:01:42.308 --> 00:01:45.428
and it swings its leg between
the two legs and catches the fall.

00:01:45.428 --> 00:01:46.793
So, that's how it walks.

00:01:46.793 --> 00:01:49.996
If you think about it, it looks
very complicated, almost organic.

00:01:49.996 --> 00:01:51.608
But why are we trying to do this?

00:01:51.608 --> 00:01:53.572
How is this biologically inspired?

00:01:53.572 --> 00:01:55.606
Let me talk about it a little bit.

00:01:55.606 --> 00:01:58.560
So, when you look at us
human beings, bipedal walking,

00:01:58.560 --> 00:02:01.049
what you're doing is
you're not really using a muscle

00:02:01.049 --> 00:02:03.896
to lift your leg and walk like a robot.
Right?

00:02:03.896 --> 00:02:07.303
What you're doing is you really swing
your leg and catch the fall,

00:02:07.303 --> 00:02:09.964
stand up again,
swing your leg and catch the fall.

00:02:09.964 --> 00:02:13.109
You're using your built-in dynamics,
the physics of your body,

00:02:13.109 --> 00:02:14.887
just like a pendulum.

00:02:14.887 --> 00:02:18.309
We call that the concept
of passive dynamic locomotion.

00:02:18.309 --> 00:02:21.469
What you're doing is, when you stand up,

00:02:21.469 --> 00:02:23.698
potential energy to kinetic energy,

00:02:23.698 --> 00:02:25.468
potential energy to kinetic energy.

00:02:25.468 --> 00:02:27.768
It's a constantly falling process.

00:02:27.768 --> 00:02:30.800
So, even though there is nothing
in nature that looks like this,

00:02:30.800 --> 00:02:32.537
really, we were inspired by biology

00:02:32.537 --> 00:02:35.954
and applying the principles of walking
to this robot.

00:02:35.954 --> 00:02:38.648
Thus it's a biologically inspired robot.

00:02:38.648 --> 00:02:41.473
What you see over here,
this is what we want to do next.

00:02:41.473 --> 00:02:44.864
We want to fold up the legs
and shoot it up for long-range motion.

00:02:44.864 --> 00:02:47.588
And it deploys legs -
it looks almost like "Star Wars" -

00:02:47.588 --> 00:02:50.469
when it lands, it absorbs
the shock and starts walking.

00:02:50.469 --> 00:02:54.161
What you see over here, this yellow thing,
this is not a death ray. (Laughter)

00:02:54.161 --> 00:02:56.529
This is just to show you
that if you have cameras

00:02:56.529 --> 00:02:58.044
or different types of sensors -

00:02:58.044 --> 00:03:00.114
because it is tall, it's 1.8 meters tall -

00:03:00.114 --> 00:03:03.215
you can see over obstacles like bushes
and those kinds of things.

00:03:03.215 --> 00:03:04.589
So we have two prototypes.

00:03:04.589 --> 00:03:07.529
The first version, in the back,
that's STriDER I.

00:03:07.529 --> 00:03:10.193
One of the problems
that we had with STriDER I -

00:03:10.193 --> 00:03:12.768
The one in front, the smaller,
is STriDER II.

00:03:12.768 --> 00:03:14.804
The problem that we had
with STriDER I is

00:03:14.804 --> 00:03:16.544
it was just too heavy in the body.

00:03:16.544 --> 00:03:19.162
We had so many motors,
you know, aligning the joints,

00:03:19.162 --> 00:03:20.514
and those kinds of things.

00:03:20.514 --> 00:03:23.486
So, we decided to synthesize
a mechanical mechanism

00:03:23.486 --> 00:03:26.498
so we could get rid of all the motors,
and with a single motor

00:03:26.498 --> 00:03:28.271
we can coordinate all the motions.

00:03:28.271 --> 00:03:32.427
It's a mechanical solution to a problem,
instead of using mechatronics.

00:03:32.540 --> 00:03:35.550
So, with this now the top body
is light enough.

00:03:35.550 --> 00:03:39.135
So, it's walking in our lab;
this was the very first successful step.

00:03:39.135 --> 00:03:41.599
It's still not perfected -
its coffee falls down -

00:03:41.599 --> 00:03:43.739
so we still have a lot of work to do.

00:03:44.083 --> 00:03:46.716
The second robot I want to talk about
is called IMPASS.

00:03:46.716 --> 00:03:47.605
It stands for

00:03:47.605 --> 00:03:51.035
Intelligent Mobility Platform
with Actuated Spoke System.

00:03:51.035 --> 00:03:53.882
So, it's a wheel-leg hybrid robot.

00:03:53.882 --> 00:03:56.020
So, think of a rimless wheel

00:03:56.020 --> 00:03:57.630
or a spoke wheel,

00:03:57.630 --> 00:04:01.003
but the spokes individually
move in and out of the hub;

00:04:01.003 --> 00:04:02.877
so, it's a wheel-leg hybrid.

00:04:02.877 --> 00:04:05.413
We are literally re-inventing
the wheel here.

00:04:05.413 --> 00:04:07.503
Let me demonstrate how it works.

00:04:07.503 --> 00:04:09.845
So, in this video we're using an approach

00:04:09.845 --> 00:04:11.974
called the reactive approach.

00:04:11.974 --> 00:04:14.788
Just simply using the tactile sensors
on the feet,

00:04:14.788 --> 00:04:17.821
it's trying to walk over
a changing terrain,

00:04:17.821 --> 00:04:20.726
a soft terrain
where it pushes down and changes.

00:04:20.726 --> 00:04:22.710
And just by the tactile information,

00:04:22.710 --> 00:04:25.742
it successfully crosses over
these type of terrain.

00:04:25.742 --> 00:04:29.100
But, when it encounters
a very extreme terrain,

00:04:29.100 --> 00:04:32.814
in this case, this obstacle
is more than three times

00:04:32.814 --> 00:04:34.951
the height of the robot,

00:04:34.951 --> 00:04:36.884
Then it switches to a deliberate mode,

00:04:36.884 --> 00:04:38.989
where it uses a laser range finder,

00:04:38.989 --> 00:04:41.961
and camera systems,
to identify the obstacle and the size,

00:04:41.961 --> 00:04:44.962
and it plans, carefully plans
the motion of the spokes

00:04:44.962 --> 00:04:47.094
and coordinates it
so that it can show this

00:04:47.094 --> 00:04:49.004
kind of very very impressive mobility.

00:04:49.004 --> 00:04:51.597
You probably haven't seen
anything like this out there.

00:04:51.597 --> 00:04:53.582
This is a very high mobility robot

00:04:53.582 --> 00:04:55.921
that we developed called IMPASS.

00:04:56.468 --> 00:05:00.422
When you drive your car,
when you steer it,

00:05:00.422 --> 00:05:02.468
you use a method called
Ackermann steering,


00:05:02.468 --> 00:05:04.508
the front wheels rotate like this.

00:05:05.339 --> 00:05:09.738
But most of the small wheeled robots
use a method called differential steering

00:05:09.738 --> 00:05:12.539
where the left and right wheel
turn in opposite directions.

00:05:12.858 --> 00:05:16.013
For IMPASS, we can do many,
many different types of motion.

00:05:16.013 --> 00:05:19.583
For example, in this case, even though
the left and right wheel is connected

00:05:19.583 --> 00:05:22.369
with a single axle rotating
at the same angle of velocity,

00:05:22.369 --> 00:05:24.462
we just simply change
the length of the spoke.

00:05:24.462 --> 00:05:27.812
It affects the diameter and then
can turn to the left and to the right.

00:05:27.812 --> 00:05:29.184
These are just some examples

00:05:29.184 --> 00:05:31.429
of the neat things
that we can do with IMPASS.

00:05:31.429 --> 00:05:33.414
This robot is called CLIMBeR:

00:05:33.414 --> 00:05:36.619
Cable-suspended Limbed Intelligent
Matching Behavior Robot.

00:05:36.619 --> 00:05:39.577
So, I've been talking to a lot
of NASA JPL scientists -

00:05:39.577 --> 00:05:41.626
at JPL they are famous
for the Mars rovers -

00:05:41.626 --> 00:05:44.001
and the scientists,
geologists always tell me

00:05:44.001 --> 00:05:46.122
that the real interesting science,

00:05:46.122 --> 00:05:48.759
the science-rich sites,
are always at the cliffs.

00:05:48.759 --> 00:05:51.271
But the current rovers cannot get there.

00:05:51.271 --> 00:05:53.904
So, inspired by that
we wanted to build a robot

00:05:53.904 --> 00:05:57.037
that can climb a structured
cliff environment.

00:05:57.037 --> 00:05:58.716
So, this is CLIMBeR.

00:05:58.716 --> 00:06:01.511
So, what it does, it has three legs.
It's difficult to see,

00:06:01.511 --> 00:06:03.584
but it has a winch
and a cable at the top -

00:06:03.584 --> 00:06:06.417
and it tries to figure out
the best place to put its foot.

00:06:06.417 --> 00:06:08.283
And then once it figures that out

00:06:08.283 --> 00:06:10.805
in real time, it calculates
the force distribution:

00:06:10.805 --> 00:06:13.285
how much force it needs
to exert to the surface

00:06:13.285 --> 00:06:15.228
so it doesn't tip and doesn't slip.

00:06:15.228 --> 00:06:17.489
Once it stabilizes that, it lifts a foot,

00:06:17.489 --> 00:06:21.016
and then with the winch
it can climb up these kinds of thing.

00:06:21.333 --> 00:06:24.405
Also for search and rescue
applications as well.

00:06:24.876 --> 00:06:28.465
This robot is called MARS:
Multi-Appendage Robotic System.

00:06:28.465 --> 00:06:30.815
Five years ago I actually
worked at NASA JPL

00:06:30.815 --> 00:06:33.053
during the summer as a faculty fellow.

00:06:33.053 --> 00:06:36.466
And they already had
a six legged robot called LEMUR.

00:06:36.466 --> 00:06:39.003
So, this is actually based on that.

00:06:39.297 --> 00:06:40.820
So, it's a hexapod robot.

00:06:40.820 --> 00:06:42.760
We developed our adaptive gait planner.

00:06:42.760 --> 00:06:45.330
We actually have a very interesting
payload on there.

00:06:45.330 --> 00:06:47.044
The students like to have fun.

00:06:47.044 --> 00:06:49.280
It shows very interesting mobility,

00:06:49.280 --> 00:06:53.456
and here you can see that it's walking
over a structured terrain.

00:06:53.456 --> 00:06:56.731
It's little bit difficult to see,
in the videos over here,

00:06:56.731 --> 00:07:00.143
it's trying to walk
on the coastal terrain, sandy area,

00:07:00.143 --> 00:07:04.887
but depending on the moisture content
or the grain size of the sand

00:07:04.887 --> 00:07:07.640
the foot's soil sinkage model changes.

00:07:07.710 --> 00:07:09.427
So, it tries to adapt its gait

00:07:09.427 --> 00:07:12.158
to successfully cross over
these kind of things.

00:07:12.158 --> 00:07:14.627
It also does some fun stuff,
as you can imagine.

00:07:14.627 --> 00:07:16.953
We get so many visitors visiting our lab.

00:07:16.953 --> 00:07:19.801
So, when the visitors come,
MARS walks up to the computer,

00:07:19.801 --> 00:07:21.985
starts typing "Hello, my name is MARS.

00:07:21.985 --> 00:07:23.266
Welcome to RoMeLa,

00:07:23.266 --> 00:07:26.361
the Robotics Mechanisms Laboratory
at Virginia Tech."

00:07:28.202 --> 00:07:30.448
This robot is an amoeba robot.

00:07:30.448 --> 00:07:33.292
Now, we don't have enough time
to go into technical details,

00:07:33.292 --> 00:07:35.934
I'll just show you some
of the experiments.

00:07:35.935 --> 00:07:38.512
So, this is some of the early
feasibility experiments.

00:07:38.512 --> 00:07:41.857
We store potential energy
to the elastic skin to make it move.

00:07:41.857 --> 00:07:45.500
Or use active tension cords
to make it move forward and backward.

00:07:45.500 --> 00:07:50.166
We've also been working with scientists
and engineers from UPenn

00:07:50.166 --> 00:07:53.765
to come up with a chemically
actuated version of this Amoeba robot.

00:07:53.775 --> 00:07:56.200
We do something to something,

00:07:56.200 --> 00:08:00.548
and just like magic, it moves. The blob.

00:08:02.086 --> 00:08:03.997
It's called ChIMERA.

00:08:04.106 --> 00:08:06.130
This robot is a very recent project.

00:08:06.130 --> 00:08:07.691
It's called RAPHaEL.

00:08:07.691 --> 00:08:10.239
Robotic Air Powered Hand
with Elastic Ligaments.

00:08:10.239 --> 00:08:14.020
There are a lot of really neat, very good
robotic hands out there in the market.

00:08:14.020 --> 00:08:17.422
The problem is they're just too expensive,
tens of thousands of dollars.

00:08:17.422 --> 00:08:20.459
So, for prosthesis applications
it's probably not too practical,

00:08:20.459 --> 00:08:22.209
because it's not affordable.

00:08:22.209 --> 00:08:25.367
We wanted to go tackle this problem
in a very different direction.

00:08:25.367 --> 00:08:28.533
Instead of using electrical motors,
electromechanical actuators,

00:08:28.533 --> 00:08:30.280
we're using compressed air.

00:08:30.280 --> 00:08:32.973
We developed these
novel actuators for joints.

00:08:32.973 --> 00:08:35.368
It is compliant.
You can actually change the force,

00:08:35.368 --> 00:08:37.490
simply just changing the air pressure.

00:08:37.490 --> 00:08:39.759
And it can actually crush
an empty soda can.

00:08:39.759 --> 00:08:43.059
It can pick up very delicate objects
like a raw egg,

00:08:43.059 --> 00:08:45.211
or in this case, a lightbulb.

00:08:45.211 --> 00:08:48.909
The best part, it took only $200 dollars
to make the first prototype.

00:08:49.946 --> 00:08:53.054
This robot is actually
a family of snake robots

00:08:53.054 --> 00:08:54.421
that we call HyDRAS,

00:08:54.421 --> 00:08:57.112
Hyper Degrees-of-freedom
Robotic Articulated Serpentine.

00:08:57.112 --> 00:09:00.331
The one that you see over here -
you can see it outdoors in the lobby

00:09:00.331 --> 00:09:03.367
we actually have a demo,
please stop by during the break time.

00:09:03.367 --> 00:09:05.494
This is a robot that can climb structures.

00:09:05.494 --> 00:09:07.488
This is a HyDRAS's arm.

00:09:07.488 --> 00:09:09.493
It's a 12 degrees of freedom robotic arm.

00:09:09.493 --> 00:09:11.724
But the cool part is the user interface.

00:09:11.724 --> 00:09:14.554
The cable over there,
that's an optical fiber.

00:09:14.554 --> 00:09:16.969
And this student,
probably the first time using it,

00:09:16.969 --> 00:09:19.166
but she can articulate
it many different ways.

00:09:19.166 --> 00:09:21.564
So, for example in Iraq,
you know, the war zone,

00:09:21.564 --> 00:09:23.038
there is roadside bombs.

00:09:23.038 --> 00:09:26.774
Currently you send these remotely
controlled vehicles that are armed.

00:09:26.774 --> 00:09:29.081
It takes really a lot of time
and it's expensive

00:09:29.081 --> 00:09:32.278
to train the operator
to operate this complex arm.

00:09:32.512 --> 00:09:34.247
In this case it's very intuitive;

00:09:34.247 --> 00:09:36.465
this student, probably
his first time using it,

00:09:36.465 --> 00:09:38.478
doing very complex manipulation tasks,

00:09:38.478 --> 00:09:42.005
picking up objects and doing manipulation,
just like that.

00:09:42.810 --> 00:09:44.100
Very intuitive.

00:09:46.066 --> 00:09:48.634
Now, this robot is currently
our star robot.

00:09:48.634 --> 00:09:51.835
We actually have a fan club
for the robot, DARwIn:

00:09:51.835 --> 00:09:54.624
Dynamic Anthropomorphic Robot
with Intelligence.

00:09:54.624 --> 00:09:58.040
As you know, we are very interested
in human walking,

00:09:58.040 --> 00:10:00.543
so we decided to build
a small humanoid robot.

00:10:00.543 --> 00:10:02.349
This was in 2004; at that time,

00:10:02.349 --> 00:10:04.326
this was something really revolutionary.

00:10:04.326 --> 00:10:06.153
This was more of a feasibility study:

00:10:06.153 --> 00:10:07.819
What kind of motors should we use?

00:10:07.819 --> 00:10:10.507
Is it even possible?
What kinds of controls should we do?

00:10:10.507 --> 00:10:12.267
So, this does not have any sensors.

00:10:12.267 --> 00:10:13.976
So, it's an open loop control.

00:10:13.976 --> 00:10:16.742
For those who probably know,
if you don't have any sensors

00:10:16.742 --> 00:10:19.699
and there are any disturbances,
you know what happens.

00:10:20.126 --> 00:10:21.950
(Laughter)

00:10:21.950 --> 00:10:24.590
So, based on that success,
the following year

00:10:24.590 --> 00:10:26.780
we did the proper mechanical design

00:10:26.780 --> 00:10:28.344
starting from kinematics.

00:10:28.344 --> 00:10:30.962
And thus, DARwIn I was born in 2005.

00:10:30.962 --> 00:10:33.603
It stands up, it walks - very impressive.

00:10:33.603 --> 00:10:36.799
However, still, as you can see,
it has a cord, umbilical cord.

00:10:36.799 --> 00:10:39.060
So, we're still using
an external power source

00:10:39.060 --> 00:10:41.264
and external computation.

00:10:41.896 --> 00:10:44.929
So, in 2006, now it's really
time to have fun.

00:10:44.929 --> 00:10:46.487
Let's give it intelligence.

00:10:46.487 --> 00:10:48.560
We give it all the computing power
it needs:

00:10:48.560 --> 00:10:50.192
a 1.5 gigahertz Pentium M chip,

00:10:50.192 --> 00:10:52.581
two FireWire cameras,
rate gyros, accelerometers,

00:10:52.581 --> 00:10:55.342
four force sensors on the foot,
lithium polymer batteries.

00:10:55.342 --> 00:10:58.957
And now DARwIn II
is completely autonomous.

00:10:58.957 --> 00:11:00.761
It is not remote controlled.

00:11:00.761 --> 00:11:03.722
There are no tethers. It looks around,
searches for the ball,

00:11:03.722 --> 00:11:07.612
looks around, searches for the ball,
and it tries to play a game of soccer,

00:11:07.612 --> 00:11:10.203
autonomously: artificial intelligence.

00:11:10.541 --> 00:11:14.188
Let's see how it does.
This was our very first trial, and...

00:11:14.188 --> 00:11:17.474
(Video): Spectators: Goal!

00:11:18.998 --> 00:11:22.171
Dennis Hong: So, there is actually
a competition called RoboCup.

00:11:22.171 --> 00:11:24.747
I don't know how many of you
have heard about RoboCup.

00:11:24.747 --> 00:11:29.087
It's an international autonomous
robot soccer competition.

00:11:29.087 --> 00:11:31.955
And the goal of RoboCup,
the actual goal is,

00:11:31.955 --> 00:11:33.883
by the year 2050

00:11:33.883 --> 00:11:37.935
we want to have full size,
autonomous humanoid robots

00:11:37.935 --> 00:11:40.902
play soccer against
the human World Cup champions

00:11:40.902 --> 00:11:42.256
and win.

00:11:42.939 --> 00:11:45.523
It's a true actual goal.
It's a very ambitious goal,

00:11:45.523 --> 00:11:47.728
but we truly believe that we can do it.

00:11:47.728 --> 00:11:49.706
So, this is last year in China.

00:11:49.706 --> 00:11:52.688
We were the very first team
in the United States that qualified

00:11:52.688 --> 00:11:54.824
in the humanoid RoboCup competition.

00:11:54.824 --> 00:11:56.945
This is this year in Austria.

00:11:56.945 --> 00:11:59.659
You're going to see the action,
three against three,

00:11:59.659 --> 00:12:01.635
completely autonomous.

00:12:01.635 --> 00:12:03.204
There you go. Yes!

00:12:04.598 --> 00:12:06.126
The robots track and they

00:12:06.126 --> 00:12:08.308
team play amongst themselves.

00:12:08.918 --> 00:12:11.320
It's very impressive.
It's really a research event

00:12:11.320 --> 00:12:15.159
packaged in a more exciting
competition event.

00:12:16.893 --> 00:12:19.060
What you see over here,
this is the beautiful

00:12:19.060 --> 00:12:20.844
Louis Vuitton Cup trophy.

00:12:20.844 --> 00:12:22.523
So, this is for the best humanoid,

00:12:22.523 --> 00:12:25.179
and we would like to bring this
for the very first time,

00:12:25.179 --> 00:12:27.419
to the United States next year,
so wish us luck.

00:12:27.419 --> 00:12:28.631
(Applause)

00:12:28.631 --> 00:12:29.842
Thank you.

00:12:32.124 --> 00:12:34.045
DARwIn also has a lot of other talents.

00:12:34.045 --> 00:12:37.876
Last year it actually conducted
the Roanoke Symphony Orchestra

00:12:37.876 --> 00:12:40.300
for the holiday concert.

00:12:40.300 --> 00:12:43.013
This is the next generation robot,
DARwIn IV,

00:12:43.013 --> 00:12:46.351
but smarter, faster, stronger.

00:12:46.351 --> 00:12:48.505
And it's trying to show off its ability:

00:12:48.505 --> 00:12:50.940
"I'm macho, I'm strong.

00:12:51.757 --> 00:12:54.146
I can also do some Jackie Chan-motion,

00:12:54.146 --> 00:12:56.000
martial art movements."

00:12:56.000 --> 00:12:57.908
(Laughter)

00:12:59.314 --> 00:13:01.299
And it walks away.
So, this is DARwIn IV.

00:13:01.299 --> 00:13:03.332
And again, you'll be able
to see it in the lobby.

00:13:03.332 --> 00:13:05.668
We truly believe this is going to be
the very first

00:13:05.668 --> 00:13:08.446
running humanoid robot
in the United States, so, stay tuned.

00:13:08.446 --> 00:13:12.306
All right. So I showed you some
of our exciting robots at work.

00:13:12.306 --> 00:13:14.456
So, what is the secret of our success?

00:13:14.456 --> 00:13:16.275
Where do we come up with these ideas?

00:13:16.275 --> 00:13:18.166
How do we develop these kinds of ideas?

00:13:18.166 --> 00:13:20.500
We win awards after awards,
year after year.

00:13:20.500 --> 00:13:23.556
We're actually running out of wall space
to put these plaques,

00:13:23.556 --> 00:13:26.970
they're staring to accumulate on the floor
hopefully we didn't loose any.

00:13:26.970 --> 00:13:29.376
These are just the awards
that we won in 2007 fall

00:13:29.376 --> 00:13:31.976
from robotics competitions
and those kinds of things.

00:13:31.976 --> 00:13:33.950
So, really, we have five secrets.

00:13:33.950 --> 00:13:36.594
First is: Where do we get inspiration?

00:13:36.594 --> 00:13:38.633
Where do we get this spark of imagination?

00:13:38.633 --> 00:13:40.555
This is a true story, my personal story.

00:13:40.555 --> 00:13:42.652
At night when I go to bed, 3 - 4 a.m.

00:13:42.652 --> 00:13:46.165
I lie down, close my eyes,
and I see these lines and circles

00:13:46.165 --> 00:13:47.975
and different shapes floating around.

00:13:47.975 --> 00:13:50.839
And they assemble, and they form
these kinds of mechanisms.

00:13:50.839 --> 00:13:52.558
And then I think, "Ah this is cool."

00:13:52.558 --> 00:13:54.747
So, right next to my bed
I keep a notebook,

00:13:54.747 --> 00:13:57.888
a journal, with a special pen
that has a light on it, LED light,

00:13:57.888 --> 00:14:00.821
because I don't want to turn on
the light and wake up my wife.

00:14:00.821 --> 00:14:04.161
So, I see this, scribble everything down,
draw things, and I go to bed.

00:14:04.161 --> 00:14:05.710
Every day in the morning,

00:14:05.710 --> 00:14:08.106
the first thing I do
before my first cup of coffee,

00:14:08.106 --> 00:14:10.293
before I brush my teeth,
I open my notebook.

00:14:10.293 --> 00:14:11.860
Many times it's empty,

00:14:11.860 --> 00:14:14.699
sometimes I have something there -
sometimes it's junk

00:14:14.699 --> 00:14:17.264
but most of the time
I can't even read my handwriting.

00:14:17.264 --> 00:14:19.718
And so, 4 in the morning,
what do you expect, right?

00:14:19.718 --> 00:14:21.611
So, I need to decipher what I wrote.

00:14:21.611 --> 00:14:24.815
But sometimes I see
this ingenious idea in there,

00:14:24.815 --> 00:14:26.585
and I have this eureka moment.

00:14:26.585 --> 00:14:29.099
I directly run to my home office,
sit at my computer,

00:14:29.099 --> 00:14:31.017
I type in the ideas, I sketch things out

00:14:31.017 --> 00:14:33.017
and I keep a database of ideas.

00:14:33.706 --> 00:14:36.182
So, when we have these
calls for proposals,

00:14:36.182 --> 00:14:40.320
I try to find a match between
my potential ideas and the problem.

00:14:40.320 --> 00:14:42.625
If there is a match
we write a research proposal,

00:14:42.625 --> 00:14:46.184
get the research funding in, and that's
how we start our research programs.

00:14:46.184 --> 00:14:48.970
But just a spark of imagination
is not good enough.

00:14:48.970 --> 00:14:51.037
How do we develop these kinds of ideas?

00:14:51.037 --> 00:14:53.820
At our lab RoMeLa, the Robotics
and Mechanisms Laboratory,

00:14:53.820 --> 00:14:56.454
we have these fantastic
brainstorming sessions.

00:14:56.454 --> 00:14:59.002
So, we gather around,
we discuss about problems

00:14:59.002 --> 00:15:02.217
and solutions to the problems
and talk about it.

00:15:02.217 --> 00:15:05.441
But before we start
we set this golden rule.

00:15:05.441 --> 00:15:06.856
The rule is:

00:15:06.856 --> 00:15:09.906
Nobody criticizes anybody's ideas.

00:15:09.906 --> 00:15:12.182
Nobody criticizes any opinion.

00:15:12.182 --> 00:15:14.906
This is important, because many times
students, they fear

00:15:14.906 --> 00:15:17.656
or they feel uncomfortable
how others might think

00:15:17.656 --> 00:15:19.765
about their opinions and thoughts.

00:15:19.765 --> 00:15:21.696
So, once you do this, it is amazing

00:15:21.696 --> 00:15:23.173
how the students open up.

00:15:23.173 --> 00:15:26.303
They have these wacky, cool,
crazy, brilliant ideas,

00:15:26.303 --> 00:15:29.778
and the whole room is just electrified
with creative energy.

00:15:29.778 --> 00:15:32.326
And this is how we develop our ideas.

00:15:32.871 --> 00:15:34.419
Well, we're running out of time.

00:15:34.419 --> 00:15:36.275
One more thing I want to talk about is,

00:15:36.275 --> 00:15:39.419
you know, just a spark of idea
and development is not good enough.

00:15:39.419 --> 00:15:41.058
There was a great TED moment,

00:15:41.058 --> 00:15:43.987
I think it was Sir Ken Robinson, was it?

00:15:43.987 --> 00:15:45.974
He gave a talk about how education

00:15:45.974 --> 00:15:48.303
and school kills creativity.

00:15:48.303 --> 00:15:50.594
Well, actually, there are
two sides to the story.

00:15:52.020 --> 00:15:54.635
So, there is only so much one can do

00:15:54.635 --> 00:15:56.521
with just ingenious ideas

00:15:56.521 --> 00:15:59.666
and creativity and good
engineering intuition.

00:15:59.666 --> 00:16:01.531
If you want to go beyond a tinkering,

00:16:01.531 --> 00:16:03.649
if you want to go beyond
a hobby of robotics

00:16:03.649 --> 00:16:07.000
and really tackle the grand challenges
of robotics

00:16:07.000 --> 00:16:09.569
through rigorous research
we need more than that.

00:16:09.569 --> 00:16:11.429
This is where school comes in.

00:16:11.526 --> 00:16:13.817
Batman, fighting against bad guys,

00:16:13.817 --> 00:16:16.340
he has his utility belt,
he has his grappling hook,

00:16:16.340 --> 00:16:18.347
he has all different kinds of gadgets.

00:16:18.347 --> 00:16:20.809
For us roboticists,
engineers and scientists,

00:16:20.809 --> 00:16:24.672
these tools, these are the courses
and classes you take in class.

00:16:24.672 --> 00:16:26.834
Math, differential equations.

00:16:26.834 --> 00:16:29.744
I have linear algebra, science, physics,

00:16:29.744 --> 00:16:32.589
even nowadays, chemistry
and biology, as you've seen.

00:16:32.589 --> 00:16:34.904
These are all the tools that we need.

00:16:34.904 --> 00:16:36.848
So, the more tools you have, for Batman,

00:16:36.848 --> 00:16:38.795
more effective at fighting the bad guys,

00:16:38.795 --> 00:16:41.478
for us, more tools to attack
these kinds of big problems.

00:16:42.510 --> 00:16:44.629
So, education is very important.

00:16:45.373 --> 00:16:48.121
Also, it's not about that,
only about that.

00:16:48.121 --> 00:16:50.255
You also have to work really, really hard.

00:16:50.255 --> 00:16:51.745
So, I always tell my students,

00:16:51.745 --> 00:16:53.830
"Work smart, then work hard."

00:16:53.830 --> 00:16:56.483
This picture in the back
this is 3 in the morning.

00:16:56.483 --> 00:16:59.090
I guarantee if you come
to your lab at 3 - 4 am

00:16:59.090 --> 00:17:00.687
we have students working there,

00:17:00.687 --> 00:17:03.860
not because I tell them to,
but because we are having too much fun.

00:17:03.860 --> 00:17:05.460
Which leads to the last topic:

00:17:05.460 --> 00:17:07.366
Do not forget to have fun.

00:17:07.366 --> 00:17:10.606
That's really the secret of our success,
we're having too much fun.

00:17:10.606 --> 00:17:14.135
I truly believe that highest productivity
comes when you're having fun,

00:17:14.135 --> 00:17:15.541
and that's what we're doing.

00:17:15.541 --> 00:17:17.122
Again, we're running out of time.

00:17:17.122 --> 00:17:20.511
Hopefully I'll have another chance
to talk to you about and introduce

00:17:20.511 --> 00:17:24.196
some other exciting robotics projects
that we didn't have time to talk about.

00:17:24.196 --> 00:17:26.272
We have a fully autonomous vehicle

00:17:26.272 --> 00:17:28.203
that can drive into urban environments.

00:17:28.203 --> 00:17:31.078
We won a half a million dollars
in the DARPA Urban Challenge.

00:17:31.078 --> 00:17:32.769
We also have the world's very first

00:17:32.769 --> 00:17:34.675
vehicle that can be driven by the blind.

00:17:34.675 --> 00:17:37.242
We call it the Blind Driver Challenge,
very exciting.

00:17:37.242 --> 00:17:40.867
And many, many other robotics projects
I want to talk about.

00:17:40.867 --> 00:17:43.532
There you go.
Go out there, read a great book.

00:17:43.532 --> 00:17:46.658
Get inspired, invent, work really hard.

00:17:46.885 --> 00:17:48.544
Stay in school.

00:17:48.544 --> 00:17:51.772
Come up with cool ideas,
I'll be happy to learn more about [them].

00:17:51.772 --> 00:17:54.106
Shoot me an email, let's talk about it.

00:17:54.106 --> 00:17:56.073
There you go. Thank you so much.

00:17:56.073 --> 00:17:58.443
(Applause)