1
00:00:00,809 --> 00:00:03,983
So the first robot
to talk about is called STriDER.

2
00:00:04,007 --> 00:00:07,476
It stands for Self-excited
Tripedal Dynamic Experimental Robot.

3
00:00:07,500 --> 00:00:11,858
It's a robot that has three legs,
which is inspired by nature.

4
00:00:12,500 --> 00:00:15,911
But have you seen anything in nature,
an animal that has three legs?

5
00:00:15,935 --> 00:00:19,805
Probably not. So why do I call this
a biologically inspired robot?

6
00:00:19,829 --> 00:00:20,991
How would it work?

7
00:00:21,015 --> 00:00:23,177
But before that,
let's look at pop culture.

8
00:00:23,201 --> 00:00:26,669
So, you know H.G. Wells's
"War of the Worlds," novel and movie.

9
00:00:26,693 --> 00:00:30,014
And what you see over here
is a very popular video game,

10
00:00:30,038 --> 00:00:33,738
and in this fiction, they describe
these alien creatures and robots

11
00:00:33,762 --> 00:00:36,024
that have three legs that terrorize Earth.

12
00:00:36,048 --> 00:00:39,583
But my robot, STriDER,
does not move like this.

13
00:00:39,607 --> 00:00:42,658
This is an actual dynamic
simulation animation.

14
00:00:42,682 --> 00:00:44,739
I'm going to show you how the robot works.

15
00:00:44,763 --> 00:00:47,155
It flips its body 180 degrees

16
00:00:47,179 --> 00:00:49,237
and it swings its leg between the two legs

17
00:00:49,261 --> 00:00:50,420
and catches the fall.

18
00:00:50,444 --> 00:00:51,606
So that's how it walks.

19
00:00:51,630 --> 00:00:54,342
But when you look at us
human beings, bipedal walking,

20
00:00:54,366 --> 00:00:55,524
what you're doing is,

21
00:00:55,548 --> 00:00:59,635
you're not really using muscle
to lift your leg and walk like a robot.

22
00:00:59,659 --> 00:01:02,996
What you're doing is,
you swing your leg and catch the fall,

23
00:01:03,020 --> 00:01:05,476
stand up again, swing your leg
and catch the fall.

24
00:01:05,500 --> 00:01:08,667
You're using your built-in dynamics,
the physics of your body,

25
00:01:08,691 --> 00:01:10,442
just like a pendulum.

26
00:01:10,466 --> 00:01:13,824
We call that the concept
of passive dynamic locomotion.

27
00:01:13,848 --> 00:01:15,986
What you're doing is, when you stand up,

28
00:01:16,010 --> 00:01:18,091
potential energy to kinetic energy,

29
00:01:18,115 --> 00:01:20,092
potential energy to kinetic energy.

30
00:01:20,116 --> 00:01:22,093
It's a constantly falling process.

31
00:01:22,117 --> 00:01:25,166
So even though there is nothing
in nature that looks like this,

32
00:01:25,190 --> 00:01:29,319
really, we're inspired by biology
and applying the principles of walking

33
00:01:29,343 --> 00:01:30,500
to this robot.

34
00:01:30,524 --> 00:01:32,501
Thus, it's a biologically inspired robot.

35
00:01:32,525 --> 00:01:34,938
What you see here,
this is what we want to do next.

36
00:01:34,962 --> 00:01:38,476
We want to fold up the legs
and shoot it up for long-range motion.

37
00:01:38,500 --> 00:01:41,294
And it deploys legs --
it looks almost like "Star Wars" --

38
00:01:41,318 --> 00:01:44,182
so when it lands, it absorbs
the shock and starts walking.

39
00:01:44,206 --> 00:01:47,411
What you see over here, this yellow thing,
this is not a death ray.

40
00:01:47,435 --> 00:01:48,447
(Laughter)

41
00:01:48,471 --> 00:01:49,630
This is just to show you

42
00:01:49,654 --> 00:01:52,341
that if you have cameras
or different types of sensors,

43
00:01:52,365 --> 00:01:53,882
because it's 1.8 meters tall,

44
00:01:53,906 --> 00:01:56,992
you can see over obstacles like bushes
and those kinds of things.

45
00:01:57,016 --> 00:01:58,363
So we have two prototypes.

46
00:01:58,387 --> 00:02:01,330
The first version,
in the back, that's STriDER I.

47
00:02:01,354 --> 00:02:03,521
The one in front,
the smaller, is STriDER II.

48
00:02:03,545 --> 00:02:07,053
The problem we had with STriDER I is,
it was just too heavy in the body.

49
00:02:07,077 --> 00:02:09,661
We had so many motors aligning the joints

50
00:02:09,685 --> 00:02:10,939
and those kinds of things.

51
00:02:10,963 --> 00:02:14,163
So we decided to synthesize
a mechanical mechanism

52
00:02:14,187 --> 00:02:17,256
so we could get rid of all the motors,
and with a single motor,

53
00:02:17,280 --> 00:02:18,957
we can coordinate all the motions.

54
00:02:18,981 --> 00:02:22,358
It's a mechanical solution to a problem,
instead of using mechatronics.

55
00:02:22,382 --> 00:02:25,794
So with this, now the top body
is lighted up; it's walking in our lab.

56
00:02:25,818 --> 00:02:27,877
This was the very first successful step.

57
00:02:27,901 --> 00:02:30,569
It's still not perfected,
its coffee falls down,

58
00:02:30,593 --> 00:02:32,575
so we still have a lot of work to do.

59
00:02:33,425 --> 00:02:36,121
The second robot I want
to talk about is called IMPASS.

60
00:02:36,145 --> 00:02:40,604
It stands for Intelligent Mobility
Platform with Actuated Spoke System.

61
00:02:40,628 --> 00:02:43,223
It's a wheel-leg hybrid robot.

62
00:02:43,247 --> 00:02:47,104
So think of a rimless wheel
or a spoke wheel,

63
00:02:47,128 --> 00:02:50,105
but the spokes individually
move in and out of the hub;

64
00:02:50,129 --> 00:02:52,216
so, it's a wheel-leg hybrid.

65
00:02:52,240 --> 00:02:54,482
We're literally reinventing
the wheel here.

66
00:02:54,506 --> 00:02:56,969
Let me demonstrate how it works.

67
00:02:56,993 --> 00:03:00,842
So in this video we're using an approach
called the reactive approach.

68
00:03:00,866 --> 00:03:03,843
Just simply using
the tactile sensors on the feet,

69
00:03:03,867 --> 00:03:06,677
it's trying to walk
over a changing terrain,

70
00:03:06,701 --> 00:03:09,401
a soft terrain where it pushes
down and changes.

71
00:03:09,425 --> 00:03:11,622
And just by the tactile information,

72
00:03:11,646 --> 00:03:14,401
it successfully crosses
over these types of terrains.

73
00:03:14,425 --> 00:03:18,230
But, when it encounters
a very extreme terrain --

74
00:03:18,254 --> 00:03:22,167
in this case, this obstacle
is more than three times the height

75
00:03:22,191 --> 00:03:23,361
of the robot --

76
00:03:23,385 --> 00:03:25,259
then it switches to a deliberate mode,

77
00:03:25,283 --> 00:03:28,097
where it uses a laser range finder
and camera systems

78
00:03:28,121 --> 00:03:29,947
to identify the obstacle and the size.

79
00:03:29,971 --> 00:03:32,955
And it carefully plans
the motion of the spokes

80
00:03:32,979 --> 00:03:36,688
and coordinates it so it can show
this very impressive mobility.

81
00:03:36,712 --> 00:03:39,353
You probably haven't seen
anything like this out there.

82
00:03:39,377 --> 00:03:43,047
This is a very high-mobility robot
that we developed called IMPASS.

83
00:03:44,613 --> 00:03:46,401
Ah, isn't that cool?

84
00:03:46,425 --> 00:03:49,719
When you drive your car,

85
00:03:49,743 --> 00:03:53,665
when you steer your car, you use
a method called Ackermann steering.

86
00:03:53,689 --> 00:03:55,540
The front wheels rotate like this.

87
00:03:55,564 --> 00:03:58,079
For most small-wheeled robots,

88
00:03:58,103 --> 00:04:00,352
they use a method
called differential steering

89
00:04:00,376 --> 00:04:03,233
where the left and right wheel
turn the opposite direction.

90
00:04:03,257 --> 00:04:06,155
For IMPASS, we can do many,
many different types of motion.

91
00:04:06,179 --> 00:04:07,448
For example, in this case,

92
00:04:07,472 --> 00:04:09,872
even though the left and right
wheels are connected

93
00:04:09,896 --> 00:04:12,662
with a single axle rotating
at the same angle of velocity,

94
00:04:12,686 --> 00:04:15,820
we simply change the length
of the spoke, it affects the diameter,

95
00:04:15,844 --> 00:04:17,931
then can turn to the left
and to the right.

96
00:04:17,955 --> 00:04:21,308
These are just some examples
of the neat things we can do with IMPASS.

97
00:04:21,332 --> 00:04:23,106
This robot is called CLIMBeR:

98
00:04:23,130 --> 00:04:26,530
Cable-suspended Limbed Intelligent
Matching Behavior Robot.

99
00:04:26,554 --> 00:04:29,681
I've been talking
to a lot of NASA JPL scientists --

100
00:04:29,705 --> 00:04:31,855
at JPL, they are famous
for the Mars rovers --

101
00:04:31,879 --> 00:04:34,250
and the scientists,
geologists always tell me

102
00:04:34,274 --> 00:04:37,401
that the real interesting science,
the science-rich sites,

103
00:04:37,425 --> 00:04:38,936
are always at the cliffs.

104
00:04:38,960 --> 00:04:40,910
But the current rovers cannot get there.

105
00:04:40,934 --> 00:04:43,417
So, inspired by that,
we wanted to build a robot

106
00:04:43,441 --> 00:04:46,558
that can climb
a structured cliff environment.

107
00:04:46,582 --> 00:04:47,806
So this is CLIMBeR.

108
00:04:47,830 --> 00:04:49,341
It has three legs.

109
00:04:49,365 --> 00:04:52,948
It's probably difficult to see, but it has
a winch and a cable at the top.

110
00:04:52,972 --> 00:04:55,553
It tries to figure out
the best place to put its foot.

111
00:04:55,577 --> 00:04:57,212
And then once it figures that out,

112
00:04:57,236 --> 00:05:00,213
in real time, it calculates
the force distribution:

113
00:05:00,237 --> 00:05:02,760
how much force it needs
to exert to the surface

114
00:05:02,784 --> 00:05:04,761
so it doesn't tip and doesn't slip.

115
00:05:04,785 --> 00:05:06,926
Once it stabilizes that, it lifts a foot,

116
00:05:06,950 --> 00:05:10,253
and then with the winch,
it can climb up these kinds of cliffs.

117
00:05:10,785 --> 00:05:13,401
Also for search and rescue
applications as well.

118
00:05:13,425 --> 00:05:15,844
Five years ago,
I actually worked at NASA JPL

119
00:05:15,868 --> 00:05:17,718
during the summer as a faculty fellow.

120
00:05:17,742 --> 00:05:21,192
And they already had
a six-legged robot called LEMUR.

121
00:05:21,216 --> 00:05:22,856
So this is actually based on that.

122
00:05:22,880 --> 00:05:24,171
This robot is called MARS:

123
00:05:24,195 --> 00:05:25,812
Multi-Appendage Robotic System.

124
00:05:25,836 --> 00:05:27,074
It's a hexapod robot.

125
00:05:27,098 --> 00:05:29,075
We developed our adaptive gait planner.

126
00:05:29,099 --> 00:05:31,624
We actually have a very interesting
payload on there.

127
00:05:31,648 --> 00:05:33,095
The students like to have fun.

128
00:05:33,119 --> 00:05:36,214
And here you can see that it's walking
over unstructured terrain.

129
00:05:36,238 --> 00:05:37,355
(Motor sound)

130
00:05:37,379 --> 00:05:40,176
It's trying to walk
on the coastal terrain, a sandy area,

131
00:05:40,200 --> 00:05:44,848
but depending on the moisture content
or the grain size of the sand,

132
00:05:44,872 --> 00:05:48,716
the foot's soil sinkage model changes,
so it tries to adapt its gait

133
00:05:48,740 --> 00:05:50,980
to successfully cross
over these kind of things.

134
00:05:51,004 --> 00:05:52,511
It also does some fun stuff.

135
00:05:52,535 --> 00:05:56,020
As you can imagine,
we get so many visitors visiting our lab.

136
00:05:56,044 --> 00:05:58,776
So when the visitors come,
MARS walks up to the computer,

137
00:05:58,800 --> 00:06:00,848
starts typing, "Hello, my name is MARS.

138
00:06:00,872 --> 00:06:02,401
Welcome to RoMeLa,

139
00:06:02,425 --> 00:06:05,038
the Robotics Mechanisms
Laboratory at Virginia Tech."

140
00:06:05,062 --> 00:06:06,532
(Laughter)

141
00:06:06,556 --> 00:06:08,681
This robot is an amoeba robot.

142
00:06:08,705 --> 00:06:11,593
Now, we don't have enough time
to go into technical details,

143
00:06:11,617 --> 00:06:13,674
I'll just show you
some of the experiments.

144
00:06:13,698 --> 00:06:16,210
These are some of the early
feasibility experiments.

145
00:06:16,234 --> 00:06:19,319
We store potential energy
to the elastic skin to make it move,

146
00:06:19,343 --> 00:06:23,146
or use active tension cords
to make it move forward and backward.

147
00:06:23,170 --> 00:06:24,327
It's called ChIMERA.

148
00:06:24,351 --> 00:06:27,363
We also have been working
with some scientists and engineers

149
00:06:27,387 --> 00:06:28,538
from UPenn

150
00:06:28,562 --> 00:06:32,395
to come up with a chemically actuated
version of this amoeba robot.

151
00:06:32,419 --> 00:06:34,302
We do something to something,

152
00:06:34,326 --> 00:06:36,994
and just like magic, it moves.

153
00:06:37,626 --> 00:06:39,208
"The Blob."

154
00:06:39,933 --> 00:06:41,710
This robot is a very recent project.

155
00:06:41,734 --> 00:06:42,885
It's called RAPHaEL:

156
00:06:42,909 --> 00:06:45,596
Robotic Air-Powered Hand
with Elastic Ligaments.

157
00:06:45,620 --> 00:06:48,619
There are a lot of really neat,
very good robotic hands

158
00:06:48,643 --> 00:06:49,806
out there on the market.

159
00:06:49,830 --> 00:06:52,010
The problem is,
they're just too expensive --

160
00:06:52,034 --> 00:06:53,439
tens of thousands of dollars.

161
00:06:53,463 --> 00:06:56,484
So for prosthesis applications
it's probably not too practical,

162
00:06:56,508 --> 00:06:57,853
because it's not affordable.

163
00:06:57,877 --> 00:07:01,351
We wanted to tackle this problem
in a very different direction.

164
00:07:01,375 --> 00:07:04,424
Instead of using electrical motors,
electromechanical actuators,

165
00:07:04,448 --> 00:07:06,204
we're using compressed air.

166
00:07:06,228 --> 00:07:09,675
We developed these novel actuators
for the joints, so it's compliant.

167
00:07:09,699 --> 00:07:11,401
You can actually change the force,

168
00:07:11,425 --> 00:07:13,251
simply just changing the air pressure.

169
00:07:13,275 --> 00:07:15,465
And it can actually crush
an empty soda can.

170
00:07:15,489 --> 00:07:18,539
It can pick up very delicate
objects like a raw egg,

171
00:07:18,563 --> 00:07:20,151
or in this case, a lightbulb.

172
00:07:20,790 --> 00:07:24,459
The best part: it took only 200 dollars
to make the first prototype.

173
00:07:25,906 --> 00:07:28,654
This robot is actually
a family of snake robots

174
00:07:28,678 --> 00:07:30,046
that we call HyDRAS,

175
00:07:30,070 --> 00:07:32,781
Hyper Degrees-of-freedom Robotic
Articulated Serpentine.

176
00:07:32,805 --> 00:07:34,976
This is a robot that can climb structures.

177
00:07:35,000 --> 00:07:37,083
This is a HyDRAS's arm.

178
00:07:37,107 --> 00:07:39,116
It's a 12-degrees-of-freedom robotic arm.

179
00:07:39,140 --> 00:07:41,488
But the cool part is the user interface.

180
00:07:41,512 --> 00:07:44,255
The cable over there,
that's an optical fiber.

181
00:07:44,279 --> 00:07:46,715
This student, it's probably
her first time using it,

182
00:07:46,739 --> 00:07:49,096
but she can articulate it
in many different ways.

183
00:07:49,120 --> 00:07:52,583
So, for example, in Iraq, the war zone,
there are roadside bombs.

184
00:07:52,607 --> 00:07:56,618
Currently, you send these remotely
controlled vehicles that are armed.

185
00:07:56,642 --> 00:08:00,181
It takes really a lot of time
and it's expensive to train the operator

186
00:08:00,205 --> 00:08:01,943
to operate this complex arm.

187
00:08:01,967 --> 00:08:03,934
In this case, it's very intuitive;

188
00:08:03,958 --> 00:08:06,226
this student, probably
his first time using it,

189
00:08:06,250 --> 00:08:08,228
is doing very complex manipulation tasks,

190
00:08:08,252 --> 00:08:11,613
picking up objects and doing
manipulation, just like that.

191
00:08:11,637 --> 00:08:12,842
Very intuitive.

192
00:08:14,765 --> 00:08:17,362
Now, this robot
is currently our star robot.

193
00:08:17,386 --> 00:08:20,371
We actually have a fan club
for the robot, DARwIn:

194
00:08:20,395 --> 00:08:23,296
Dynamic Anthropomorphic
Robot with Intelligence.

195
00:08:23,320 --> 00:08:27,203
As you know, we're very interested
in human walking,

196
00:08:27,227 --> 00:08:29,419
so we decided to build
a small humanoid robot.

197
00:08:29,443 --> 00:08:31,205
This was in 2004; at that time,

198
00:08:31,229 --> 00:08:33,499
this was something really,
really revolutionary.

199
00:08:33,523 --> 00:08:35,324
This was more of a feasibility study:

200
00:08:35,348 --> 00:08:37,999
What kind of motors should we use?
Is it even possible?

201
00:08:38,023 --> 00:08:39,757
What kinds of controls should we do?

202
00:08:39,781 --> 00:08:43,301
This does not have any sensors,
so it's an open-loop control.

203
00:08:43,325 --> 00:08:46,095
For those who probably know,
if you don't have any sensors

204
00:08:46,119 --> 00:08:48,692
and there's any disturbances,
you know what happens.

205
00:08:48,716 --> 00:08:50,829
(Laughter)

206
00:08:50,853 --> 00:08:56,077
Based on that success, the following year
we did the proper mechanical design,

207
00:08:56,101 --> 00:08:57,325
starting from kinematics.

208
00:08:57,349 --> 00:09:00,259
And thus, DARwIn I was born in 2005.

209
00:09:00,283 --> 00:09:02,594
It stands up, it walks -- very impressive.

210
00:09:02,618 --> 00:09:05,975
However, still, as you can see,
it has a cord, an umbilical cord.

211
00:09:05,999 --> 00:09:08,171
So we're still using
an external power source

212
00:09:08,195 --> 00:09:09,794
and external computation.

213
00:09:10,742 --> 00:09:14,053
So in 2006, now it's really
time to have fun.

214
00:09:14,077 --> 00:09:15,536
Let's give it intelligence.

215
00:09:15,560 --> 00:09:17,681
We give it all the computing
power it needs:

216
00:09:17,705 --> 00:09:20,235
a 1.5 gigahertz Pentium M chip,
two FireWire cameras,

217
00:09:20,259 --> 00:09:23,105
rate gyros, accelerometers,
four forced sensors on the foot,

218
00:09:23,129 --> 00:09:24,479
lithium polymer batteries --

219
00:09:24,503 --> 00:09:27,653
and now DARwIn II
is completely autonomous.

220
00:09:28,146 --> 00:09:30,837
It is not remote controlled.
There's no tethers.

221
00:09:30,861 --> 00:09:34,594
It looks around, searches for the ball ...
looks around, searches for the ball,

222
00:09:34,618 --> 00:09:39,547
and it tries to play a game of soccer
autonomously -- artificial intelligence.

223
00:09:39,571 --> 00:09:40,725
Let's see how it does.

224
00:09:40,749 --> 00:09:42,696
This was our very first trial, and ...

225
00:09:42,720 --> 00:09:47,087
(Video) Spectators: Goal!

226
00:09:48,238 --> 00:09:51,188
Dennis Hong: There is actually
a competition called RoboCup.

227
00:09:51,212 --> 00:09:53,799
I don't know how many of you
have heard about RoboCup.

228
00:09:53,823 --> 00:09:58,197
It's an international autonomous
robot soccer competition.

229
00:09:58,221 --> 00:10:00,969
And the actual goal of RoboCup is,

230
00:10:00,993 --> 00:10:03,186
by the year 2050,

231
00:10:03,210 --> 00:10:06,941
we want to have full-size,
autonomous humanoid robots

232
00:10:06,965 --> 00:10:10,039
play soccer against the human
World Cup champions

233
00:10:10,063 --> 00:10:11,216
and win.

234
00:10:11,240 --> 00:10:12,272
(Laughter)

235
00:10:12,296 --> 00:10:13,514
It's a true, actual goal.

236
00:10:13,538 --> 00:10:17,435
It's a very ambitious goal,
but we truly believe we can do it.

237
00:10:17,459 --> 00:10:19,193
This is last year in China.

238
00:10:19,217 --> 00:10:22,230
We were the very first team
in the United States that qualified

239
00:10:22,254 --> 00:10:24,054
in the humanoid RoboCup competition.

240
00:10:24,078 --> 00:10:26,258
This is this year in Austria.

241
00:10:26,282 --> 00:10:28,849
You're going to see the action
is three against three,

242
00:10:28,873 --> 00:10:30,064
completely autonomous.

243
00:10:30,088 --> 00:10:31,179
(Video) (Crowd groans)

244
00:10:31,203 --> 00:10:32,533
DH: There you go. Yes!

245
00:10:33,331 --> 00:10:37,473
The robots track and they team-play
amongst themselves.

246
00:10:37,934 --> 00:10:39,085
It's very impressive.

247
00:10:39,109 --> 00:10:40,585
It's really a research event,

248
00:10:40,609 --> 00:10:44,849
packaged in a more exciting
competition event.

249
00:10:44,873 --> 00:10:48,401
What you see here is the beautiful
Louis Vuitton Cup trophy.

250
00:10:48,425 --> 00:10:49,944
This is for the best humanoid.

251
00:10:49,968 --> 00:10:53,637
We'd like to bring this, for the first
time, to the United States next year,

252
00:10:53,661 --> 00:10:54,816
so wish us luck.

253
00:10:54,840 --> 00:10:55,876
(Applause)

254
00:10:55,900 --> 00:10:57,052
Thank you.

255
00:10:57,076 --> 00:10:59,157
(Applause)

256
00:10:59,181 --> 00:11:01,249
DARwIn also has a lot of other talents.

257
00:11:01,273 --> 00:11:04,989
Last year, it actually conducted
the Roanoke Symphony Orchestra

258
00:11:05,013 --> 00:11:07,401
for the holiday concert.

259
00:11:07,425 --> 00:11:10,401
This is the next generation
robot, DARwIn IV,

260
00:11:10,425 --> 00:11:13,401
much smarter, faster, stronger.

261
00:11:13,425 --> 00:11:15,402
And it's trying to show off its ability:

262
00:11:15,426 --> 00:11:17,223
"I'm macho, I'm strong."

263
00:11:17,247 --> 00:11:18,691
(Laughter)

264
00:11:18,715 --> 00:11:22,940
"I can also do some Jackie Chan-motion,
martial art movements."

265
00:11:22,964 --> 00:11:24,976
(Laughter)

266
00:11:26,425 --> 00:11:28,360
And it walks away. So this is DARwIn IV.

267
00:11:28,384 --> 00:11:30,519
Again, you'll be able
to see it in the lobby.

268
00:11:30,543 --> 00:11:33,735
We truly believe this will be
the very first running humanoid robot

269
00:11:33,759 --> 00:11:34,910
in the United States.

270
00:11:34,934 --> 00:11:36,088
So stay tuned.

271
00:11:36,112 --> 00:11:39,113
All right. So I showed you
some of our exciting robots at work.

272
00:11:39,137 --> 00:11:41,325
So, what is the secret of our success?

273
00:11:41,349 --> 00:11:43,166
Where do we come up with these ideas?

274
00:11:43,190 --> 00:11:45,083
How do we develop these kinds of ideas?

275
00:11:45,107 --> 00:11:46,897
We have a fully autonomous vehicle

276
00:11:46,921 --> 00:11:48,797
that can drive into urban environments.

277
00:11:48,821 --> 00:11:51,720
We won a half a million dollars
in the DARPA Urban Challenge.

278
00:11:51,744 --> 00:11:55,401
We also have the world's very first
vehicle that can be driven by the blind.

279
00:11:55,425 --> 00:11:57,949
We call it the Blind Driver
Challenge, very exciting.

280
00:11:57,973 --> 00:12:01,145
And many, many other robotics
projects I want to talk about.

281
00:12:01,169 --> 00:12:03,806
These are just the awards
that we won in 2007 fall

282
00:12:03,830 --> 00:12:06,401
from robotics competitions
and those kinds of things.

283
00:12:06,425 --> 00:12:08,318
So really, we have five secrets.

284
00:12:08,342 --> 00:12:10,778
First is: Where do we get inspiration?

285
00:12:10,802 --> 00:12:12,881
Where do we get this spark of imagination?

286
00:12:12,905 --> 00:12:14,926
This is a true story, my personal story.

287
00:12:14,950 --> 00:12:17,727
At night, when I go to bed,
at three, four in the morning,

288
00:12:17,751 --> 00:12:20,640
I lie down, close my eyes,
and I see these lines and circles

289
00:12:20,664 --> 00:12:22,458
and different shapes floating around.

290
00:12:22,482 --> 00:12:25,245
And they assemble, and they form
these kinds of mechanisms.

291
00:12:25,269 --> 00:12:26,877
And I think, "Ah, this is cool."

292
00:12:26,901 --> 00:12:29,778
So right next to my bed
I keep a notebook, a journal,

293
00:12:29,802 --> 00:12:32,158
with a special pen
that has an LED light on it,

294
00:12:32,182 --> 00:12:35,219
because I don't want to turn on the light
and wake up my wife.

295
00:12:35,243 --> 00:12:38,512
So I see this, scribble everything down,
draw things, and go to bed.

296
00:12:38,536 --> 00:12:40,926
Every day in the morning,
the first thing I do,

297
00:12:40,950 --> 00:12:43,557
before my first cup of coffee,
before I brush my teeth,

298
00:12:43,581 --> 00:12:44,743
I open my notebook.

299
00:12:44,767 --> 00:12:47,430
Many times it's empty;
sometimes I have something there.

300
00:12:47,454 --> 00:12:49,520
If something's there, sometimes it's junk.

301
00:12:49,544 --> 00:12:51,901
But most of the time,
I can't read my handwriting.

302
00:12:51,925 --> 00:12:54,249
Four in the morning --
what do you expect, right?

303
00:12:54,273 --> 00:12:56,177
So I need to decipher what I wrote.

304
00:12:56,201 --> 00:12:59,401
But sometimes I see
this ingenious idea in there,

305
00:12:59,425 --> 00:13:00,910
and I have this eureka moment.

306
00:13:00,934 --> 00:13:03,450
I directly run to my home office,
sit at my computer,

307
00:13:03,474 --> 00:13:05,451
I type in the ideas, I sketch things out

308
00:13:05,475 --> 00:13:07,366
and I keep a database of ideas.

309
00:13:08,028 --> 00:13:10,389
So when we have these calls for proposals,

310
00:13:10,413 --> 00:13:13,983
I try to find a match
between my potential ideas

311
00:13:14,007 --> 00:13:15,167
and the problem.

312
00:13:15,191 --> 00:13:17,508
If there's a match,
we write a research proposal,

313
00:13:17,532 --> 00:13:18,879
get the research funding in,

314
00:13:18,903 --> 00:13:21,063
and that's how we start
our research programs.

315
00:13:21,087 --> 00:13:23,580
But just a spark of imagination
is not good enough.

316
00:13:23,604 --> 00:13:25,484
How do we develop these kinds of ideas?

317
00:13:25,508 --> 00:13:28,288
At our lab RoMeLa, the Robotics
and Mechanisms Laboratory,

318
00:13:28,312 --> 00:13:30,924
we have these fantastic
brainstorming sessions.

319
00:13:30,948 --> 00:13:35,283
So we gather around, we discuss problems
and solutions and talk about it.

320
00:13:35,307 --> 00:13:38,284
But before we start,
we set this golden rule.

321
00:13:38,308 --> 00:13:39,957
The rule is:

322
00:13:39,981 --> 00:13:42,958
nobody criticizes anybody's ideas.

323
00:13:42,982 --> 00:13:44,987
Nobody criticizes any opinion.

324
00:13:45,425 --> 00:13:48,972
This is important, because many times,
students fear or feel uncomfortable

325
00:13:48,996 --> 00:13:52,265
about how others might think
about their opinions and thoughts.

326
00:13:52,289 --> 00:13:56,079
So once you do this, it is amazing
how the students open up.

327
00:13:56,103 --> 00:13:59,401
They have these wacky, cool,
crazy, brilliant ideas,

328
00:13:59,425 --> 00:14:02,854
and the whole room is just electrified
with creative energy.

329
00:14:02,878 --> 00:14:05,151
And this is how we develop our ideas.

330
00:14:05,738 --> 00:14:07,295
Well, we're running out of time.

331
00:14:07,319 --> 00:14:09,221
One more thing I want to talk about is,

332
00:14:09,245 --> 00:14:12,401
you know, just a spark of idea
and development is not good enough.

333
00:14:12,425 --> 00:14:17,136
There was a great TED moment --
I think it was Sir Ken Robinson, was it?

334
00:14:17,160 --> 00:14:21,598
He gave a talk about how education
and school kill creativity.

335
00:14:21,622 --> 00:14:24,401
Well, actually,
there's two sides to the story.

336
00:14:24,425 --> 00:14:29,638
So there is only so much one can do
with just ingenious ideas

337
00:14:29,662 --> 00:14:32,678
and creativity
and good engineering intuition.

338
00:14:32,702 --> 00:14:34,496
If you want to go beyond a tinkering,

339
00:14:34,520 --> 00:14:36,670
if you want to go
beyond a hobby of robotics

340
00:14:36,694 --> 00:14:40,135
and really tackle
the grand challenges of robotics

341
00:14:40,159 --> 00:14:41,401
through rigorous research,

342
00:14:41,425 --> 00:14:42,580
we need more than that.

343
00:14:42,604 --> 00:14:44,618
This is where school comes in.

344
00:14:44,642 --> 00:14:47,136
Batman, fighting against the bad guys,

345
00:14:47,160 --> 00:14:49,553
he has his utility belt,
he has his grappling hook,

346
00:14:49,577 --> 00:14:51,401
he has all different kinds of gadgets.

347
00:14:51,425 --> 00:14:53,838
For us roboticists,
engineers and scientists,

348
00:14:53,862 --> 00:14:58,019
these tools are the courses
and classes you take in class.

349
00:14:58,043 --> 00:15:00,045
Math, differential equations.

350
00:15:00,069 --> 00:15:02,809
I have linear algebra, science, physics --

351
00:15:02,833 --> 00:15:05,801
even, nowadays, chemistry
and biology, as you've seen.

352
00:15:05,825 --> 00:15:07,583
These are all the tools we need.

353
00:15:07,607 --> 00:15:09,483
So the more tools you have, for Batman,

354
00:15:09,507 --> 00:15:11,484
more effective at fighting the bad guys,

355
00:15:11,508 --> 00:15:14,414
for us, more tools to attack
these kinds of big problems.

356
00:15:15,582 --> 00:15:17,447
So education is very important.

357
00:15:18,568 --> 00:15:21,165
Also -- it's not only about that.

358
00:15:21,189 --> 00:15:23,257
You also have to work really, really hard.

359
00:15:23,281 --> 00:15:24,734
So I always tell my students,

360
00:15:24,758 --> 00:15:26,809
"Work smart, then work hard."

361
00:15:26,833 --> 00:15:29,557
This picture in the back --
this is three in the morning.

362
00:15:29,581 --> 00:15:31,910
I guarantee if you come
to our lab at 3, 4am,

363
00:15:31,934 --> 00:15:33,490
we have students working there,

364
00:15:33,514 --> 00:15:36,723
not because I tell them to,
but because we are having too much fun.

365
00:15:36,747 --> 00:15:38,401
Which leads to the last topic:

366
00:15:38,425 --> 00:15:40,294
do not forget to have fun.

367
00:15:40,318 --> 00:15:43,531
That's really the secret of our success,
we're having too much fun.

368
00:15:43,555 --> 00:15:47,021
I truly believe that highest productivity
comes when you're having fun,

369
00:15:47,045 --> 00:15:48,401
and that's what we're doing.

370
00:15:48,425 --> 00:15:49,579
And there you go.

371
00:15:49,603 --> 00:15:50,770
Thank you so much.

372
00:15:50,794 --> 00:15:54,896
(Applause)