1
00:00:05,651 --> 00:00:08,779
I think we found
a really fun spot in robotics

2
00:00:08,863 --> 00:00:10,781
that hasn't been explored
very much before.

3
00:00:14,368 --> 00:00:15,953
It's a good day to throw a robot, I'd say.

4
00:00:16,037 --> 00:00:17,163
I think so.

5
00:00:17,580 --> 00:00:19,623
[Morgan] So, right now,
we're doing a couple of things.

6
00:00:19,707 --> 00:00:22,251
One is just the--
Every morning you gotta get him up

7
00:00:22,334 --> 00:00:24,920
and make sure he can do his calisthenics.

8
00:00:25,004 --> 00:00:26,589
Make sure he's feeling fit and fierce.

9
00:00:27,131 --> 00:00:28,132
Yeah, head's coming up.

10
00:00:28,215 --> 00:00:29,300
[roboticist 2] Head's coming up.

11
00:00:30,134 --> 00:00:31,302
Good morning.

12
00:00:45,983 --> 00:00:48,569
I grew up super, like,
super bookish, super nerdy.

13
00:00:48,652 --> 00:00:51,113
I grew up in a little town in rural Utah.

14
00:00:52,073 --> 00:00:53,824
I got a really good ACT score,
and someone said,

15
00:00:53,908 --> 00:00:55,743
"You could probably go to Harvard
if you wanted to."

16
00:00:55,826 --> 00:00:57,495
Then I looked up the ACT scores like,
"Maybe."

17
00:00:58,329 --> 00:00:59,455
And then I got in.

18
00:01:00,039 --> 00:01:02,124
I ran into somebody
who was doing his PhD program.

19
00:01:02,208 --> 00:01:03,209
I was like, "So, what do you do?"

20
00:01:03,292 --> 00:01:06,212
And he's like, "Well, you know,
I build robots, I 3-D print these robots,

21
00:01:06,295 --> 00:01:08,089
and then I use them to test algorithms

22
00:01:08,172 --> 00:01:09,882
for how robots could collaborate
with each other."

23
00:01:09,965 --> 00:01:12,176
And I was like, "Wow, that's super cool.
That's awesome.

24
00:01:12,259 --> 00:01:14,261
It's amazing that you get the opportunity
to do that stuff."

25
00:01:14,345 --> 00:01:17,264
I was like,
"Oh, there's no reason I can't do that."

26
00:01:18,599 --> 00:01:20,684
So this is the pose
that I'm thinking of modifying

27
00:01:20,768 --> 00:01:22,603
so he's kinda got a little bit of tension

28
00:01:22,686 --> 00:01:24,939
when he's actually in the, like,
in the blocks.

29
00:01:25,022 --> 00:01:28,484
My role at Disney is
as a Research and Development Imagineer.

30
00:01:28,567 --> 00:01:32,446
And so, my job is to look at what
technologies are available and also think,

31
00:01:32,530 --> 00:01:35,282
"What is the new technology
that we could make

32
00:01:35,366 --> 00:01:39,245
that would allow us to do something that
hasn't been done before in the world?"

33
00:01:40,454 --> 00:01:43,958
Since I started at Disney, my main focus
has been this project called Stuntronics,

34
00:01:44,041 --> 00:01:46,419
which, at first, wasn't something
that was especially well-defined.

35
00:01:46,502 --> 00:01:47,795
We just had this idea of,

36
00:01:47,878 --> 00:01:50,965
"How can we control things
while they're spinning through the air?"

37
00:01:52,717 --> 00:01:55,761
Parks right now, you come to them
and you see all of our characters

38
00:01:55,845 --> 00:01:58,013
who you're used to seeing
doing amazing action sequences.

39
00:02:00,516 --> 00:02:02,852
What we'd like to do is get these
characters off the ground and moving.

40
00:02:06,981 --> 00:02:10,234
The vision is,
"Can we deliver these big action beats?"

41
00:02:10,317 --> 00:02:13,195
Previously, they've only expected
to have those delivered digitally, right?

42
00:02:13,279 --> 00:02:14,780
Like the magic of CGI.

43
00:02:14,864 --> 00:02:18,117
It would be wonderful to imagine
that as a real tactile experience.

44
00:02:19,160 --> 00:02:20,786
What we'd really love,
at the end of the day,

45
00:02:20,870 --> 00:02:23,456
is for people to be like,
"How did they do that?"

46
00:02:23,539 --> 00:02:25,332
These robots,
they don't look good close-up.

47
00:02:25,416 --> 00:02:28,878
They're built to last
and to drop from 65 feet in the air.

48
00:02:29,378 --> 00:02:30,629
It's always been the idea

49
00:02:30,713 --> 00:02:33,340
that we would have some kind
of a hero performer up front,

50
00:02:33,424 --> 00:02:35,801
kind of leading into the show,
and then, somehow, change out

51
00:02:35,885 --> 00:02:37,470
and do this really amazing thing.

52
00:02:37,553 --> 00:02:40,514
The really difficult problem
is trying to get an acrobatic robot

53
00:02:40,598 --> 00:02:42,641
to fly through the air and do
all the cool things you want it to do.

54
00:02:42,725 --> 00:02:44,060
So in order to help solve that problem,

55
00:02:44,143 --> 00:02:46,312
we need to look beyond what we can see
with the human eye.

56
00:02:46,395 --> 00:02:48,814
So we have a bunch
of high-speed cameras set up.

57
00:02:48,898 --> 00:02:52,485
We also have the robot
who's keeping track of servo positions,

58
00:02:52,568 --> 00:02:54,111
where its arm is at every time,

59
00:02:54,528 --> 00:02:57,031
his rotation and acceleration
and velocity.

60
00:02:57,114 --> 00:03:00,326
Now, if you take all that data and you
look at it all at once after-the-fact,

61
00:03:00,409 --> 00:03:01,911
you can kinda start to tease apart,

62
00:03:01,994 --> 00:03:06,457
"Okay, what are the little variations
that are producing the effects

63
00:03:06,540 --> 00:03:08,292
that I either want or don't want to see?"

64
00:03:08,918 --> 00:03:12,129
So, I'm just looking at stuff like,
"Is he tilting from side to side?

65
00:03:12,213 --> 00:03:14,840
Has he got a little bit of a degree or two
of twist?"

66
00:03:15,466 --> 00:03:17,009
When you're a robotics researcher,

67
00:03:17,093 --> 00:03:18,803
you're touching everything
at the beginning phase.

68
00:03:18,886 --> 00:03:20,638
You gotta write a little code,
make a little circuit,

69
00:03:20,721 --> 00:03:22,556
you gotta build something mechanically.

70
00:03:22,640 --> 00:03:24,183
And so that was super fun

71
00:03:24,266 --> 00:03:26,936
because when you have a little bit
of all those skill sets,

72
00:03:27,019 --> 00:03:29,355
what you can make is something that's new.

73
00:03:29,438 --> 00:03:32,983
So, these are our first three, like,
kind of, full-scale figures

74
00:03:33,067 --> 00:03:35,361
when we started to think, "Could we
actually scale this up to person-size?"

75
00:03:35,444 --> 00:03:38,447
So the first,
which we then added one joint to,

76
00:03:38,531 --> 00:03:41,242
that was just to give us an idea of,
"Can we throw something?

77
00:03:41,325 --> 00:03:43,994
Can we kinda tuck a little bit
and get an acceleration?"

78
00:03:44,078 --> 00:03:45,162
Turned out you could.

79
00:03:45,246 --> 00:03:47,415
So we moved on from this beautiful thing

80
00:03:47,498 --> 00:03:49,917
to something
that was actually made of metal.

81
00:03:50,000 --> 00:03:53,170
This guy, he can fold in two places.
We threw this one a lot.

82
00:03:53,254 --> 00:03:54,380
Learned a lot about the physics

83
00:03:54,463 --> 00:03:56,549
of how you actually throw something
through the air.

84
00:03:56,632 --> 00:04:00,678
This is our first, admittedly
fairly crude, human-form robot.

85
00:04:00,761 --> 00:04:02,471
We liked this one.
He was a good performer.

86
00:04:02,555 --> 00:04:04,640
These prototypes,
even though they're really simple,

87
00:04:04,724 --> 00:04:07,309
they were super important
for our development process.

88
00:04:07,393 --> 00:04:12,440
You knew we were onto something good
when this three-stick thing hit the ground

89
00:04:12,523 --> 00:04:14,567
and I was like,
"Oh!" Like, I felt bad for it.

90
00:04:14,650 --> 00:04:17,028
So even a crude prototype
can reveal something like that.

91
00:04:17,111 --> 00:04:19,155
So, as a roboticist,
I find that super exciting

92
00:04:19,238 --> 00:04:22,450
because now
we can create the illusion of life

93
00:04:22,533 --> 00:04:24,410
in a way that's really unique and new.

94
00:04:26,120 --> 00:04:29,623
I don't know, I was thinking maybe
the torso's kicked back more or something.

95
00:04:29,707 --> 00:04:31,333
-[roboticist 2] So, more of an arch?
-More of an arch.

96
00:04:32,293 --> 00:04:34,086
Those are nice. Arched back.

97
00:04:36,422 --> 00:04:38,716
[Morgan] We can control, okay,
when do you see this robot?

98
00:04:38,799 --> 00:04:40,926
Where is it coming from?
What is it landing on?

99
00:04:41,010 --> 00:04:44,138
How far away are the people?
And what's the story that's framing it?

100
00:04:44,597 --> 00:04:47,266
I can build something
that's a new robot to the world

101
00:04:47,349 --> 00:04:49,351
because I'm in a place
where I'm telling stories.

102
00:04:49,435 --> 00:04:51,395
And then I wrap a story around it,

103
00:04:51,479 --> 00:04:53,814
and suddenly, it's even more awesome
than it would otherwise be.

104
00:04:55,483 --> 00:04:57,818
My favorite film
of the Walt Disney Company

105
00:04:57,902 --> 00:05:01,155
is actually our first feature-length one,
Snow White and the Seven Dwarfs.

106
00:05:01,530 --> 00:05:05,409
Walt was relentless in his drive to make
animation come from what it had been

107
00:05:05,493 --> 00:05:08,579
to being something that could be beautiful
and moving and still funny.

108
00:05:09,372 --> 00:05:13,167
And he employed new techniques,
really talented people,

109
00:05:13,250 --> 00:05:15,461
impressive technical innovations
to create that

110
00:05:15,544 --> 00:05:18,172
and also, though,
just this driving vision of like,

111
00:05:18,255 --> 00:05:21,008
"Let's do something
that nobody thinks can be done."

112
00:05:22,385 --> 00:05:24,887
I think that's something that,
hopefully, lets people see

113
00:05:24,970 --> 00:05:27,181
that robotics can do these amazing things,

114
00:05:27,264 --> 00:05:29,141
even if they're not practical yet
in the real world.

115
00:05:29,225 --> 00:05:32,728
And I hope we're continuing that legacy
in projects like Stuntronics.

116
00:05:33,229 --> 00:05:34,230
It's nice to come to a place

117
00:05:34,313 --> 00:05:36,732
where your job is to do something
that's never been done before

118
00:05:36,816 --> 00:05:39,443
and that people haven't seen
and that excites people.

119
00:05:39,527 --> 00:05:42,613
And so your job
is to do something magical.