0:00:00.000,0:00:03.000
So, the first robot to talk about is called STriDER.

0:00:03.000,0:00:05.000
It stands for Self-excited

0:00:05.000,0:00:07.000
Tripedal Dynamic Experimental Robot.

0:00:07.000,0:00:09.000
It's a robot that has three legs,

0:00:09.000,0:00:12.000
which is inspired by nature.

0:00:12.000,0:00:14.000
But have you seen anything in nature,

0:00:14.000,0:00:16.000
an animal that has three legs?

0:00:16.000,0:00:18.000
Probably not. So, why do I call this

0:00:18.000,0:00:20.000
a biologically inspired robot? How would it work?

0:00:20.000,0:00:23.000
But before that, let's look at pop culture.

0:00:23.000,0:00:26.000
So, you know H.G. Wells' "War of the Worlds," novel and movie.

0:00:26.000,0:00:28.000
And what you see over here is a very popular

0:00:28.000,0:00:30.000
video game,

0:00:30.000,0:00:33.000
and in this fiction they describe these alien creatures that

0:00:33.000,0:00:35.000
are robots that have three legs that terrorize Earth.

0:00:35.000,0:00:39.000
But my robot, STriDER, does not move like this.

0:00:39.000,0:00:42.000
So, this is an actual dynamic simulation animation.

0:00:42.000,0:00:44.000
I'm just going to show you how the robot works.

0:00:44.000,0:00:47.000
It flips its body 180 degrees

0:00:47.000,0:00:50.000
and it swings its leg between the two legs and catches the fall.

0:00:50.000,0:00:52.000
So, that's how it walks. But when you look at us

0:00:52.000,0:00:54.000
human being, bipedal walking,

0:00:54.000,0:00:56.000
what you're doing is you're not really using a muscle

0:00:56.000,0:00:59.000
to lift your leg and walk like a robot. Right?

0:00:59.000,0:01:02.000
What you're doing is you really swing your leg and catch the fall,

0:01:02.000,0:01:05.000
stand up again, swing your leg and catch the fall.

0:01:05.000,0:01:08.000
You're using your built-in dynamics, the physics of your body,

0:01:08.000,0:01:10.000
just like a pendulum.

0:01:10.000,0:01:14.000
We call that the concept of passive dynamic locomotion.

0:01:14.000,0:01:16.000
What you're doing is, when you stand up,

0:01:16.000,0:01:18.000
potential energy to kinetic energy,

0:01:18.000,0:01:20.000
potential energy to kinetic energy.

0:01:20.000,0:01:22.000
It's a constantly falling process.

0:01:22.000,0:01:25.000
So, even though there is nothing in nature that looks like this,

0:01:25.000,0:01:27.000
really, we were inspired by biology

0:01:27.000,0:01:29.000
and applying the principles of walking

0:01:29.000,0:01:32.000
to this robot. Thus it's a biologically inspired robot.

0:01:32.000,0:01:34.000
What you see over here, this is what we want to do next.

0:01:34.000,0:01:38.000
We want to fold up the legs and shoot it up for long-range motion.

0:01:38.000,0:01:41.000
And it deploys legs -- it looks almost like "Star Wars" --

0:01:41.000,0:01:44.000
when it lands, it absorbs the shock and starts walking.

0:01:44.000,0:01:47.000
What you see over here, this yellow thing, this is not a death ray. (Laughter)

0:01:47.000,0:01:49.000
This is just to show you that if you have cameras

0:01:49.000,0:01:51.000
or different types of sensors --

0:01:51.000,0:01:53.000
because it is tall, it's 1.8 meters tall --

0:01:53.000,0:01:56.000
you can see over obstacles like bushes and those kinds of things.

0:01:56.000,0:01:58.000
So we have two prototypes.

0:01:58.000,0:02:01.000
The first version, in the back, that's STriDER I.

0:02:01.000,0:02:03.000
The one in front, the smaller, is STriDER II.

0:02:03.000,0:02:05.000
The problem that we had with STriDER I is

0:02:05.000,0:02:08.000
it was just too heavy in the body. We had so many motors,

0:02:08.000,0:02:10.000
you know, aligning the joints, and those kinds of things.

0:02:10.000,0:02:14.000
So, we decided to synthesize a mechanical mechanism

0:02:14.000,0:02:17.000
so we could get rid of all the motors, and with a single motor

0:02:17.000,0:02:19.000
we can coordinate all the motions.

0:02:19.000,0:02:22.000
It's a mechanical solution to a problem, instead of using mechatronics.

0:02:22.000,0:02:25.000
So, with this now the top body is light enough. So, it's walking in our lab;

0:02:25.000,0:02:28.000
this was the very first successful step.

0:02:28.000,0:02:30.000
It's still not perfected -- its coffee falls down --

0:02:30.000,0:02:33.000
so we still have a lot of work to do.

0:02:33.000,0:02:36.000
The second robot I want to talk about is called IMPASS.

0:02:36.000,0:02:40.000
It stands for Intelligent Mobility Platform with Actuated Spoke System.

0:02:40.000,0:02:43.000
So, it's a wheel-leg hybrid robot.

0:02:43.000,0:02:45.000
So, think of a rimless wheel

0:02:45.000,0:02:47.000
or a spoke wheel,

0:02:47.000,0:02:50.000
but the spokes individually move in and out of the hub;

0:02:50.000,0:02:52.000
so, it's a wheel-leg hybrid.

0:02:52.000,0:02:54.000
We are literally re-inventing the wheel here.

0:02:54.000,0:02:57.000
Let me demonstrate how it works.

0:02:57.000,0:02:59.000
So, in this video we're using an approach

0:02:59.000,0:03:01.000
called the reactive approach.

0:03:01.000,0:03:04.000
Just simply using the tactile sensors on the feet,

0:03:04.000,0:03:06.000
it's trying to walk over a changing terrain,

0:03:06.000,0:03:09.000
a soft terrain where it pushes down and changes.

0:03:09.000,0:03:11.000
And just by the tactile information,

0:03:11.000,0:03:14.000
it successfully crosses over these type of terrain.

0:03:14.000,0:03:18.000
But, when it encounters a very extreme terrain,

0:03:18.000,0:03:21.000
in this case, this obstacle is more than three times

0:03:21.000,0:03:23.000
the height of the robot,

0:03:23.000,0:03:25.000
Then it switches to a deliberate mode,

0:03:25.000,0:03:27.000
where it uses a laser range finder,

0:03:27.000,0:03:29.000
and camera systems, to identify the obstacle and the size,

0:03:29.000,0:03:32.000
and it plans, carefully plans the motion of the spokes

0:03:32.000,0:03:34.000
and coordinates it so that it can show this

0:03:34.000,0:03:36.000
kind of very very impressive mobility.

0:03:36.000,0:03:38.000
You probably haven't seen anything like this out there.

0:03:38.000,0:03:41.000
This is a very high mobility robot

0:03:41.000,0:03:44.000
that we developed called IMPASS.

0:03:44.000,0:03:46.000
Ah, isn't that cool?

0:03:46.000,0:03:49.000
When you drive your car,

0:03:49.000,0:03:51.000
when you steer your car, you use a method

0:03:51.000,0:03:53.000
called Ackermann steering.

0:03:53.000,0:03:55.000
The front wheels rotate like this.

0:03:55.000,0:03:58.000
For most small wheeled robots,

0:03:58.000,0:04:00.000
they use a method called differential steering

0:04:00.000,0:04:03.000
where the left and right wheel turns the opposite direction.

0:04:03.000,0:04:06.000
For IMPASS, we can do many, many different types of motion.

0:04:06.000,0:04:09.000
For example, in this case, even though the left and right wheel is connected

0:04:09.000,0:04:11.000
with a single axle rotating at the same angle of velocity.

0:04:11.000,0:04:14.000
We just simply change the length of the spoke.

0:04:14.000,0:04:16.000
It affects the diameter and then can turn to the left, turn to the right.

0:04:16.000,0:04:18.000
So, these are just some examples of the neat things

0:04:18.000,0:04:21.000
that we can do with IMPASS.

0:04:21.000,0:04:23.000
This robot is called CLIMBeR:

0:04:23.000,0:04:26.000
Cable-suspended Limbed Intelligent Matching Behavior Robot.

0:04:26.000,0:04:29.000
So, I've been talking to a lot of NASA JPL scientists --

0:04:29.000,0:04:31.000
at JPL they are famous for the Mars rovers --

0:04:31.000,0:04:33.000
and the scientists, geologists always tell me

0:04:33.000,0:04:36.000
that the real interesting science,

0:04:36.000,0:04:39.000
the science-rich sites, are always at the cliffs.

0:04:39.000,0:04:41.000
But the current rovers cannot get there.

0:04:41.000,0:04:43.000
So, inspired by that we wanted to build a robot

0:04:43.000,0:04:46.000
that can climb a structured cliff environment.

0:04:46.000,0:04:48.000
So, this is CLIMBeR.

0:04:48.000,0:04:50.000
So, what it does, it has three legs. It's probably difficult to see,

0:04:50.000,0:04:53.000
but it has a winch and a cable at the top --

0:04:53.000,0:04:55.000
and it tries to figure out the best place to put its foot.

0:04:55.000,0:04:57.000
And then once it figures that out

0:04:57.000,0:05:00.000
in real time, it calculates the force distribution:

0:05:00.000,0:05:03.000
how much force it needs to exert to the surface

0:05:03.000,0:05:05.000
so it doesn't tip and doesn't slip.

0:05:05.000,0:05:07.000
Once it stabilizes that, it lifts a foot,

0:05:07.000,0:05:11.000
and then with the winch it can climb up these kinds of thing.

0:05:11.000,0:05:13.000
Also for search and rescue applications as well.

0:05:13.000,0:05:15.000
Five years ago I actually worked at NASA JPL

0:05:15.000,0:05:17.000
during the summer as a faculty fellow.

0:05:17.000,0:05:21.000
And they already had a six legged robot called LEMUR.

0:05:21.000,0:05:24.000
So, this is actually based on that. This robot is called MARS:

0:05:24.000,0:05:27.000
Multi-Appendage Robotic System. So, it's a hexapod robot.

0:05:27.000,0:05:29.000
We developed our adaptive gait planner.

0:05:29.000,0:05:31.000
We actually have a very interesting payload on there.

0:05:31.000,0:05:33.000
The students like to have fun. And here you can see that it's

0:05:33.000,0:05:36.000
walking over unstructured terrain.

0:05:36.000,0:05:38.000
It's trying to walk on the coarse terrain,

0:05:38.000,0:05:40.000
sandy area,

0:05:40.000,0:05:45.000
but depending on the moisture content or the grain size of the sand

0:05:45.000,0:05:47.000
the foot's soil sinkage model changes.

0:05:47.000,0:05:51.000
So, it tries to adapt its gait to successfully cross over these kind of things.

0:05:51.000,0:05:53.000
And also, it does some fun stuff, as can imagine.

0:05:53.000,0:05:56.000
We get so many visitors visiting our lab.

0:05:56.000,0:05:58.000
So, when the visitors come, MARS walks up to the computer,

0:05:58.000,0:06:00.000
starts typing "Hello, my name is MARS."

0:06:00.000,0:06:02.000
Welcome to RoMeLa,

0:06:02.000,0:06:06.000
the Robotics Mechanisms Laboratory at Virginia Tech.

0:06:06.000,0:06:08.000
This robot is an amoeba robot.

0:06:08.000,0:06:11.000
Now, we don't have enough time to go into technical details,

0:06:11.000,0:06:13.000
I'll just show you some of the experiments.

0:06:13.000,0:06:15.000
So, this is some of the early feasibility experiments.

0:06:15.000,0:06:19.000
We store potential energy to the elastic skin to make it move.

0:06:19.000,0:06:21.000
Or use an active tension cords to make it move

0:06:21.000,0:06:24.000
forward and backward. It's called ChIMERA.

0:06:24.000,0:06:26.000
We also have been working with some scientists

0:06:26.000,0:06:28.000
and engineers from UPenn

0:06:28.000,0:06:30.000
to come up with a chemically actuated version

0:06:30.000,0:06:32.000
of this amoeba robot.

0:06:32.000,0:06:34.000
We do something to something

0:06:34.000,0:06:40.000
And just like magic, it moves. The blob.

0:06:40.000,0:06:42.000
This robot is a very recent project. It's called RAPHaEL.

0:06:42.000,0:06:45.000
Robotic Air Powered Hand with Elastic Ligaments.

0:06:45.000,0:06:49.000
There are a lot of really neat, very good robotic hands out there in the market.

0:06:49.000,0:06:53.000
The problem is they're just too expensive, tens of thousands of dollars.

0:06:53.000,0:06:55.000
So, for prosthesis applications it's probably not too practical,

0:06:55.000,0:06:57.000
because it's not affordable.

0:06:57.000,0:07:01.000
We wanted to go tackle this problem in a very different direction.

0:07:01.000,0:07:04.000
Instead of using electrical motors, electromechanical actuators,

0:07:04.000,0:07:06.000
we're using compressed air.

0:07:06.000,0:07:08.000
We developed these novel actuators for joints.

0:07:08.000,0:07:11.000
It is compliant. You can actually change the force,

0:07:11.000,0:07:13.000
simply just changing the air pressure.

0:07:13.000,0:07:15.000
And it can actually crush an empty soda can.

0:07:15.000,0:07:18.000
It can pick up very delicate objects like a raw egg,

0:07:18.000,0:07:21.000
or in this case, a lightbulb.

0:07:21.000,0:07:25.000
The best part, it took only $200 dollars to make the first prototype.

0:07:25.000,0:07:28.000
This robot is actually a family of snake robots

0:07:28.000,0:07:30.000
that we call HyDRAS,

0:07:30.000,0:07:32.000
Hyper Degrees-of-freedom Robotic Articulated Serpentine.

0:07:32.000,0:07:35.000
This is a robot that can climb structures.

0:07:35.000,0:07:37.000
This is a HyDRAS's arm.

0:07:37.000,0:07:39.000
It's a 12 degrees of freedom robotic arm.

0:07:39.000,0:07:41.000
But the cool part is the user interface.

0:07:41.000,0:07:44.000
The cable over there, that's an optical fiber.

0:07:44.000,0:07:46.000
And this student, probably the first time using it,

0:07:46.000,0:07:48.000
but she can articulate it many different ways.

0:07:48.000,0:07:51.000
So, for example in Iraq, you know, the war zone,

0:07:51.000,0:07:53.000
there is roadside bombs. Currently you send these

0:07:53.000,0:07:56.000
remotely controlled vehicles that are armed.

0:07:56.000,0:07:58.000
It takes really a lot of time and it's expensive

0:07:58.000,0:08:02.000
to train the operator to operate this complex arm.

0:08:02.000,0:08:04.000
In this case it's very intuitive;

0:08:04.000,0:08:08.000
this student, probably his first time using it, doing very complex manipulation tasks,

0:08:08.000,0:08:10.000
picking up objects and doing manipulation,

0:08:10.000,0:08:13.000
just like that. Very intuitive.

0:08:15.000,0:08:17.000
Now, this robot is currently our star robot.

0:08:17.000,0:08:20.000
We actually have a fan club for the robot, DARwIn:

0:08:20.000,0:08:23.000
Dynamic Anthropomorphic Robot with Intelligence.

0:08:23.000,0:08:25.000
As you know, we are very interested in

0:08:25.000,0:08:27.000
humanoid robot, human walking,

0:08:27.000,0:08:29.000
so we decided to build a small humanoid robot.

0:08:29.000,0:08:31.000
This was in 2004; at that time,

0:08:31.000,0:08:33.000
this was something really, really revolutionary.

0:08:33.000,0:08:35.000
This was more of a feasibility study:

0:08:35.000,0:08:37.000
What kind of motors should we use?

0:08:37.000,0:08:39.000
Is it even possible? What kinds of controls should we do?

0:08:39.000,0:08:41.000
So, this does not have any sensors.

0:08:41.000,0:08:43.000
So, it's an open loop control.

0:08:43.000,0:08:45.000
For those who probably know, if you don't have any sensors

0:08:45.000,0:08:47.000
and there are any disturbances, you know what happens.

0:08:50.000,0:08:51.000
(Laughter)

0:08:51.000,0:08:53.000
So, based on that success, the following year

0:08:53.000,0:08:56.000
we did the proper mechanical design

0:08:56.000,0:08:58.000
starting from kinematics.

0:08:58.000,0:09:00.000
And thus, DARwIn I was born in 2005.

0:09:00.000,0:09:02.000
It stands up, it walks -- very impressive.

0:09:02.000,0:09:04.000
However, still, as you can see,

0:09:04.000,0:09:08.000
it has a cord, umbilical cord. So, we're still using an external power source

0:09:08.000,0:09:10.000
and external computation.

0:09:10.000,0:09:14.000
So, in 2006, now it's really time to have fun.

0:09:14.000,0:09:17.000
Let's give it intelligence. We give it all the computing power it needs:

0:09:17.000,0:09:19.000
a 1.5 gigahertz Pentium M chip,

0:09:19.000,0:09:21.000
two FireWire cameras, rate gyros, accelerometers,

0:09:21.000,0:09:24.000
four force sensors on the foot, lithium polymer batteries.

0:09:24.000,0:09:28.000
And now DARwIn II is completely autonomous.

0:09:28.000,0:09:30.000
It is not remote controlled.

0:09:30.000,0:09:33.000
There are no tethers. It looks around, searches for the ball,

0:09:33.000,0:09:36.000
looks around, searches for the ball, and it tries to play a game of soccer,

0:09:36.000,0:09:39.000
autonomously: artificial intelligence.

0:09:39.000,0:09:42.000
Let's see how it does. This was our very first trial,

0:09:42.000,0:09:47.000
and... Spectators (Video): Goal!

0:09:48.000,0:09:51.000
Dennis Hong: So, there is actually a competition called RoboCup.

0:09:51.000,0:09:53.000
I don't know how many of you have heard about RoboCup.

0:09:53.000,0:09:58.000
It's an international autonomous robot soccer competition.

0:09:58.000,0:10:01.000
And the goal of RoboCup, the actual goal is,

0:10:01.000,0:10:03.000
by the year 2050

0:10:03.000,0:10:06.000
we want to have full size, autonomous humanoid robots

0:10:06.000,0:10:10.000
play soccer against the human World Cup champions

0:10:10.000,0:10:12.000
and win.

0:10:12.000,0:10:14.000
It's a true actual goal. It's a very ambitious goal,

0:10:14.000,0:10:16.000
but we truly believe that we can do it.

0:10:16.000,0:10:19.000
So, this is last year in China.

0:10:19.000,0:10:21.000
We were the very first team in the United States that qualified

0:10:21.000,0:10:23.000
in the humanoid RoboCup competition.

0:10:23.000,0:10:26.000
This is this year in Austria.

0:10:26.000,0:10:28.000
You're going to see the action, three against three,

0:10:28.000,0:10:30.000
completely autonomous.

0:10:30.000,0:10:32.000
There you go. Yes!

0:10:33.000,0:10:35.000
The robots track and they

0:10:35.000,0:10:38.000
team play amongst themselves.

0:10:38.000,0:10:40.000
It's very impressive. It's really a research event

0:10:40.000,0:10:44.000
packaged in a more exciting competition event.

0:10:44.000,0:10:46.000
What you see over here, this is the beautiful

0:10:46.000,0:10:48.000
Louis Vuitton Cup trophy.

0:10:48.000,0:10:50.000
So, this is for the best humanoid,

0:10:50.000,0:10:52.000
and we would like to bring this for the very first time, to the United States

0:10:52.000,0:10:54.000
next year, so wish us luck.

0:10:54.000,0:10:56.000
(Applause)

0:10:56.000,0:10:59.000
Thank you.

0:10:59.000,0:11:01.000
DARwIn also has a lot of other talents.

0:11:01.000,0:11:04.000
Last year it actually conducted the Roanoke Symphony Orchestra

0:11:04.000,0:11:07.000
for the holiday concert.

0:11:07.000,0:11:10.000
This is the next generation robot, DARwIn IV,

0:11:10.000,0:11:13.000
but smarter, faster, stronger.

0:11:13.000,0:11:15.000
And it's trying to show off its ability:

0:11:15.000,0:11:18.000
"I'm macho, I'm strong.

0:11:18.000,0:11:21.000
I can also do some Jackie Chan-motion,

0:11:21.000,0:11:24.000
martial art movements."

0:11:24.000,0:11:26.000
(Laughter)

0:11:26.000,0:11:28.000
And it walks away. So, this is DARwIn IV.

0:11:28.000,0:11:30.000
And again, you'll be able to see it in the lobby.

0:11:30.000,0:11:32.000
We truly believe this is going to be the very first running

0:11:32.000,0:11:35.000
humanoid robot in the United States. So, stay tuned.

0:11:35.000,0:11:38.000
All right. So I showed you some of our exciting robots at work.

0:11:38.000,0:11:41.000
So, what is the secret of our success?

0:11:41.000,0:11:43.000
Where do we come up with these ideas?

0:11:43.000,0:11:45.000
How do we develop these kinds of ideas?

0:11:45.000,0:11:47.000
We have a fully autonomous vehicle

0:11:47.000,0:11:49.000
that can drive into urban environments. We won a half a million dollars

0:11:49.000,0:11:51.000
in the DARPA Urban Challenge.

0:11:51.000,0:11:53.000
We also have the world's very first

0:11:53.000,0:11:55.000
vehicle that can be driven by the blind.

0:11:55.000,0:11:57.000
We call it the Blind Driver Challenge, very exciting.

0:11:57.000,0:12:01.000
And many, many other robotics projects I want to talk about.

0:12:01.000,0:12:03.000
These are just the awards that we won in 2007 fall

0:12:03.000,0:12:06.000
from robotics competitions and those kinds of things.

0:12:06.000,0:12:08.000
So, really, we have five secrets.

0:12:08.000,0:12:10.000
First is: Where do we get inspiration?

0:12:10.000,0:12:12.000
Where do we get this spark of imagination?

0:12:12.000,0:12:15.000
This is a true story, my personal story.

0:12:15.000,0:12:17.000
At night when I go to bed, 3 - 4 a.m. in the morning,

0:12:17.000,0:12:20.000
I lie down, close my eyes, and I see these lines and circles

0:12:20.000,0:12:22.000
and different shapes floating around.

0:12:22.000,0:12:25.000
And they assemble, and they form these kinds of mechanisms.

0:12:25.000,0:12:27.000
And then I think, "Ah this is cool."

0:12:27.000,0:12:29.000
So, right next to my bed I keep a notebook,

0:12:29.000,0:12:32.000
a journal, with a special pen that has a light on it, LED light,

0:12:32.000,0:12:34.000
because I don't want to turn on the light and wake up my wife.

0:12:34.000,0:12:36.000
So, I see this, scribble everything down, draw things,

0:12:36.000,0:12:38.000
and I go to bed.

0:12:38.000,0:12:40.000
Every day in the morning,

0:12:40.000,0:12:42.000
the first thing I do before my first cup of coffee,

0:12:42.000,0:12:44.000
before I brush my teeth, I open my notebook.

0:12:44.000,0:12:46.000
Many times it's empty,

0:12:46.000,0:12:48.000
sometimes I have something there -- if something's there, sometimes it's junk --

0:12:48.000,0:12:51.000
but most of the time I can't even read my handwriting.

0:12:51.000,0:12:54.000
And so, 4 am in the morning, what do you expect, right?

0:12:54.000,0:12:56.000
So, I need to decipher what I wrote.

0:12:56.000,0:12:59.000
But sometimes I see this ingenious idea in there,

0:12:59.000,0:13:01.000
and I have this eureka moment.

0:13:01.000,0:13:03.000
I directly run to my home office, sit at my computer,

0:13:03.000,0:13:05.000
I type in the ideas, I sketch things out

0:13:05.000,0:13:08.000
and I keep a database of ideas.

0:13:08.000,0:13:10.000
So, when we have these calls for proposals,

0:13:10.000,0:13:12.000
I try to find a match between my

0:13:12.000,0:13:14.000
potential ideas

0:13:14.000,0:13:16.000
and the problem. If there is a match we write a research proposal,

0:13:16.000,0:13:20.000
get the research funding in, and that's how we start our research programs.

0:13:20.000,0:13:23.000
But just a spark of imagination is not good enough.

0:13:23.000,0:13:25.000
How do we develop these kinds of ideas?

0:13:25.000,0:13:28.000
At our lab RoMeLa, the Robotics and Mechanisms Laboratory,

0:13:28.000,0:13:31.000
we have these fantastic brainstorming sessions.

0:13:31.000,0:13:33.000
So, we gather around, we discuss about problems

0:13:33.000,0:13:35.000
and social problems and talk about it.

0:13:35.000,0:13:38.000
But before we start we set this golden rule.

0:13:38.000,0:13:40.000
The rule is:

0:13:40.000,0:13:43.000
Nobody criticizes anybody's ideas.

0:13:43.000,0:13:45.000
Nobody criticizes any opinion.

0:13:45.000,0:13:47.000
This is important, because many times students, they fear

0:13:47.000,0:13:50.000
or they feel uncomfortable how others might think

0:13:50.000,0:13:52.000
about their opinions and thoughts.

0:13:52.000,0:13:54.000
So, once you do this, it is amazing

0:13:54.000,0:13:56.000
how the students open up.

0:13:56.000,0:13:59.000
They have these wacky, cool, crazy, brilliant ideas, and

0:13:59.000,0:14:02.000
the whole room is just electrified with creative energy.

0:14:02.000,0:14:05.000
And this is how we develop our ideas.

0:14:05.000,0:14:08.000
Well, we're running out of time. One more thing I want to talk about is,

0:14:08.000,0:14:12.000
you know, just a spark of idea and development is not good enough.

0:14:12.000,0:14:14.000
There was a great TED moment,

0:14:14.000,0:14:17.000
I think it was Sir Ken Robinson, was it?

0:14:17.000,0:14:19.000
He gave a talk about how education

0:14:19.000,0:14:21.000
and school kills creativity.

0:14:21.000,0:14:24.000
Well, actually, there are two sides to the story.

0:14:24.000,0:14:27.000
So, there is only so much one can do

0:14:27.000,0:14:29.000
with just ingenious ideas

0:14:29.000,0:14:32.000
and creativity and good engineering intuition.

0:14:32.000,0:14:34.000
If you want to go beyond a tinkering,

0:14:34.000,0:14:36.000
if you want to go beyond a hobby of robotics

0:14:36.000,0:14:39.000
and really tackle the grand challenges of robotics

0:14:39.000,0:14:41.000
through rigorous research

0:14:41.000,0:14:44.000
we need more than that. This is where school comes in.

0:14:44.000,0:14:47.000
Batman, fighting against bad guys,

0:14:47.000,0:14:49.000
he has his utility belt, he has his grappling hook,

0:14:49.000,0:14:51.000
he has all different kinds of gadgets.

0:14:51.000,0:14:53.000
For us roboticists, engineers and scientists,

0:14:53.000,0:14:58.000
these tools, these are the courses and classes you take in class.

0:14:58.000,0:15:00.000
Math, differential equations.

0:15:00.000,0:15:02.000
I have linear algebra, science, physics,

0:15:02.000,0:15:05.000
even nowadays, chemistry and biology, as you've seen.

0:15:05.000,0:15:07.000
These are all the tools that we need.

0:15:07.000,0:15:09.000
So, the more tools you have, for Batman,

0:15:09.000,0:15:11.000
more effective at fighting the bad guys,

0:15:11.000,0:15:15.000
for us, more tools to attack these kinds of big problems.

0:15:15.000,0:15:18.000
So, education is very important.

0:15:18.000,0:15:20.000
Also, it's not about that,

0:15:20.000,0:15:22.000
only about that. You also have to work really, really hard.

0:15:22.000,0:15:24.000
So, I always tell my students,

0:15:24.000,0:15:26.000
"Work smart, then work hard."

0:15:26.000,0:15:29.000
This picture in the back this is 3 a.m. in the morning.

0:15:29.000,0:15:31.000
I guarantee if you come to your lab at 3 - 4 am

0:15:31.000,0:15:33.000
we have students working there,

0:15:33.000,0:15:36.000
not because I tell them to, but because we are having too much fun.

0:15:36.000,0:15:38.000
Which leads to the last topic:

0:15:38.000,0:15:40.000
Do not forget to have fun.

0:15:40.000,0:15:43.000
That's really the secret of our success, we're having too much fun.

0:15:43.000,0:15:46.000
I truly believe that highest productivity comes when you're having fun,

0:15:46.000,0:15:48.000
and that's what we're doing.

0:15:48.000,0:15:50.000
There you go. Thank you so much.

0:15:50.000,0:15:55.000
(Applause)