The Future of Flying Robots | Vijay Kumar | TED Talks

755,090 views ใƒป 2015-11-04

TED


์•„๋ž˜ ์˜๋ฌธ์ž๋ง‰์„ ๋”๋ธ”ํด๋ฆญํ•˜์‹œ๋ฉด ์˜์ƒ์ด ์žฌ์ƒ๋ฉ๋‹ˆ๋‹ค.

๋ฒˆ์—ญ: Jihyeon J. Kim ๊ฒ€ํ† : Gemma Lee
00:13
In my lab, we build autonomous aerial robots
0
13280
3656
์ œ ์—ฐ๊ตฌ์‹ค์€ ์—ฌ๊ธฐ ๋‚ ์•„๋‹ค๋‹ˆ๋Š” ๊ฒƒ์ฒ˜๋Ÿผ ์ž๋™ ๋น„ํ–‰ ๋กœ๋ด‡์„ ๋งŒ๋“ญ๋‹ˆ๋‹ค.
00:16
like the one you see flying here.
1
16960
1880
00:20
Unlike the commercially available drones that you can buy today,
2
20720
3696
์˜ค๋Š˜๋‚  ๊ตฌ๋งคํ•  ์ˆ˜ ์žˆ๋Š” ์ƒ์—…์šฉ ๋“œ๋ก ๊ณผ๋Š” ๋‹ฌ๋ฆฌ
00:24
this robot doesn't have any GPS on board.
3
24440
2640
์ด ๋กœ๋ด‡์€ GPS๋ฅผ ํƒ‘์žฌํ•˜๊ณ  ์žˆ์ง€ ์•Š์Šต๋‹ˆ๋‹ค.
00:28
So without GPS,
4
28160
1216
GPS์—†์ด๋Š”
00:29
it's hard for robots like this to determine their position.
5
29400
3280
์ด๋Ÿฐ ๋กœ๋ด‡์ด ์ž์‹ ์˜ ์œ„์น˜๋ฅผ ํŒŒ์•…ํ•˜๊ธฐ ํž˜๋“ญ๋‹ˆ๋‹ค.
00:34
This robot uses onboard sensors, cameras and laser scanners,
6
34240
4736
์ด ๋กœ๋ด‡์€ ์ž์ฒด ์„ผ์„œ์™€ ์นด๋ฉ”๋ผ, ๋ ˆ์ด์ € ์Šค์บ๋„ˆ๋ฅผ ๊ฐ€์ง€๊ณ 
์ฃผ๋ณ€ ํ™˜๊ฒฝ์„ ํƒ์ƒ‰ํ•ฉ๋‹ˆ๋‹ค.
00:39
to scan the environment.
7
39000
1696
00:40
It detects features from the environment,
8
40720
3056
ํ™˜๊ฒฝ์˜ ํŠน์ด์‚ฌํ•ญ๋“ค์„ ํƒ์ง€ํ•˜๊ณ 
00:43
and it determines where it is relative to those features,
9
43800
2736
๊ทธ ์‚ฌํ•ญ๋“ค์ด ๊ด€๋ จ์žˆ๋Š” ๊ณณ์—์„œ
00:46
using a method of triangulation.
10
46560
2136
์‚ผ๊ฐ์ธก๋Ÿ‰๋ฒ•์„ ์ด์šฉํ•ด์„œ ๊ฒฐ์ •ํ•ฉ๋‹ˆ๋‹ค.
00:48
And then it can assemble all these features into a map,
11
48720
3456
๊ทธ๋Ÿฌ๊ณ  ๋‚˜์„œ ์ด ํŠน์ง•๋“ค์„ ๋’ค์— ๋ณด์‹œ๋Š” ๊ฒƒ ๊ฐ™์€
00:52
like you see behind me.
12
52200
1736
์ง€๋„๋กœ ์กฐํ•ฉํ•  ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค.
00:53
And this map then allows the robot to understand where the obstacles are
13
53960
3936
์ด ์ง€๋„๋Š” ๋กœ๋ด‡์ด ์žฅ์• ๋ฌผ์˜ ์œ„์น˜๋ฅผ ์•Œ์•„์„œ
00:57
and navigate in a collision-free manner.
14
57920
2720
์ถฉ๋Œ์—†์ด ๋‚ ์•„๋‹ค๋‹ ์ˆ˜ ์žˆ๊ฒŒ ํ•ฉ๋‹ˆ๋‹ค.
01:01
What I want to show you next
15
61160
2096
๋‹ค์Œ์— ๋ณด์—ฌ ๋“œ๋ฆฌ๊ณ  ์‹ถ์€ ๊ฒƒ์€
01:03
is a set of experiments we did inside our laboratory,
16
63280
3216
์—ฐ๊ตฌ์‹ค์—์„œ ์ง„ํ–‰ํ–ˆ๋˜ ๋ช‡ ๊ฐ€์ง€ ์‹คํ—˜์ธ๋ฐ
01:06
where this robot was able to go for longer distances.
17
66520
3480
์ด ๋กœ๋ด‡์ด ์žฅ๊ฑฐ๋ฆฌ ๋น„ํ–‰์„ ํ•  ์ˆ˜ ์žˆ๋Š” ๊ฒ๋‹ˆ๋‹ค.
01:10
So here you'll see, on the top right, what the robot sees with the camera.
18
70400
5016
๋ณด์‹œ๋ฉด ์˜ค๋ฅธ์ชฝ ๋งจ ์œ„์— ๋กœ๋ด‡์ด ์นด๋ฉ”๋ผ๋กœ ๋ณด๋Š” ๊ฒƒ์ž…๋‹ˆ๋‹ค.
01:15
And on the main screen --
19
75440
1216
์ฃผ ํ™”๋ฉด์—๋Š” ์ด๊ฑด 4๋ฐฐ์†์œผ๋กœ ๋Œ๋ฆฐ ๊ฒ๋‹ˆ๋‹ค.
01:16
and of course this is sped up by a factor of four --
20
76680
2456
01:19
on the main screen you'll see the map that it's building.
21
79160
2667
์ฃผ ํ™”๋ฉด์— ์ง€๋„๋ฅผ ๋งŒ๋“œ๋Š” ๊ฒƒ์ด ๋ณด์ž…๋‹ˆ๋‹ค.
01:21
So this is a high-resolution map of the corridor around our laboratory.
22
81851
4285
์ด๊ฒƒ์€ ์ €ํฌ ์—ฐ๊ตฌ์‹ค ๋ณต๋„์˜ ๊ณ ํ™”์งˆ ์ง€๋„์ž…๋‹ˆ๋‹ค.
01:26
And in a minute you'll see it enter our lab,
23
86160
2336
์ด๊ฒƒ์ด ๊ณง ์—ฐ๊ตฌ์‹ค๋กœ ๋“ค์–ด์˜ค๋Š”๊ฒŒ ๋ณด์ผ ๊ฒ๋‹ˆ๋‹ค.
01:28
which is recognizable by the clutter that you see.
24
88520
2856
์–ด์ง€๋Ÿฝ๊ฒŒ ๋„๋ฆฐ ๋ฌผ๊ฑด๋“ค์„ ๋ณด๊ณ  ์•Œ ์ˆ˜ ์žˆ์œผ์‹œ์ฃ .
01:31
(Laughter)
25
91400
1016
(์›ƒ์Œ)
01:32
But the main point I want to convey to you
26
92440
2007
์—ฌ๋Ÿฌ๋ถ„๊ป˜ ์•Œ๋ ค๋“œ๋ฆด ์ค‘์š”ํ•œ ์ ์€
01:34
is that these robots are capable of building high-resolution maps
27
94472
3584
์ด ๋กœ๋ด‡๋“ค์ด 5cm ํ•ด์ƒ๋„์˜ ๊ณ ํ™”์งˆ ์ง€๋„๋ฅผ ๋งŒ๋“ค์–ด์„œ
01:38
at five centimeters resolution,
28
98080
2496
01:40
allowing somebody who is outside the lab, or outside the building
29
100600
4176
์—ฐ๊ตฌ์‹ค์ด๋‚˜ ๊ฑด๋ฌผ ๋ฐ–์— ์žˆ๋Š” ์‚ฌ๋žŒ์ด
01:44
to deploy these without actually going inside,
30
104800
3216
์•ˆ์— ๋“ค์–ด๊ฐ€์ง€๋„ ์•Š๊ณ  ๋กœ๋ด‡์„ ๋ณด๋‚ด๊ณ 
๊ฑด๋ฌผ ์•ˆ์—์„œ ๋ฒŒ์–ด์ง€๋Š” ์ผ์„ ์•Œ์•„๋‚ด๊ฒŒ ํ•ด ์ค๋‹ˆ๋‹ค.
01:48
and trying to infer what happens inside the building.
31
108040
3760
01:52
Now there's one problem with robots like this.
32
112400
2240
์ด๋Ÿฐ ๋กœ๋ด‡์— ํ•œ ๊ฐ€์ง€ ๋ฌธ์ œ๊ฐ€ ์žˆ์Šต๋‹ˆ๋‹ค.
01:55
The first problem is it's pretty big.
33
115600
2200
์ฒซ ๋ฒˆ์งธ ๋ฌธ์ œ๋Š” ์ด๊ฒƒ์ด ๋งค์šฐ ํฝ๋‹ˆ๋‹ค.
01:58
Because it's big, it's heavy.
34
118120
1680
ํฌ๋‹ค๋ณด๋‹ˆ ๋ฌด๊ฒ์Šต๋‹ˆ๋‹ค.
02:00
And these robots consume about 100 watts per pound.
35
120640
3040
์ด ๋กœ๋ด‡๋“ค์€ 1ํŒŒ์šด๋“œ ๋‹น 100์™€ํŠธ๋ฅผ ์†Œ๋ชจํ•ฉ๋‹ˆ๋‹ค.
02:04
And this makes for a very short mission life.
36
124360
2280
๊ทธ๋ž˜์„œ ์ž„๋ฌด์ˆ˜ํ–‰ ์‹œ๊ฐ„์ด ๋งค์šฐ ์งง์Šต๋‹ˆ๋‹ค.
02:08
The second problem
37
128000
1456
๋‘ ๋ฒˆ์งธ ๋ฌธ์ œ๋Š”
02:09
is that these robots have onboard sensors that end up being very expensive --
38
129480
3896
์ด ๋กœ๋ด‡๋“ค์ด ๋งค์šฐ ๋น„์‹ผ ๋ฐ
02:13
a laser scanner, a camera and the processors.
39
133400
3440
๋‚ด์žฅ ์„ผ์„œ, ๋ ˆ์ด์ € ์Šค์บ๋„ˆ, ์นด๋ฉ”๋ผ์™€ ์ฒ˜๋ฆฌ์žฅ์น˜๊ฐ€ ์žˆ์Šต๋‹ˆ๋‹ค.
02:17
That drives up the cost of this robot.
40
137280
3040
๊ทธ๊ฒƒ ๋•Œ๋ฌธ์— ์ด ๋กœ๋ด‡๊ฐ€๊ฒฉ์ด ์น˜์†Ÿ์Šต๋‹ˆ๋‹ค.
02:21
So we asked ourselves a question:
41
141440
2656
๊ทธ๋ž˜์„œ ์ด๋Ÿฐ ์งˆ๋ฌธ์„ ํ•ด ๋ดค์Šต๋‹ˆ๋‹ค.
02:24
what consumer product can you buy in an electronics store
42
144120
3776
์ „์ž์ œํ’ˆ ๋งค์žฅ์˜ ์–ด๋–ค ์†Œ๋น„์žฌ๊ฐ€
02:27
that is inexpensive, that's lightweight, that has sensing onboard and computation?
43
147920
6280
์ €๋ ดํ•˜๊ณ , ๊ฐ€๋ฒผ์šฐ๋ฉฐ ๋‚ด์žฅ ์„ผ์„œ์™€ ์ฒ˜๋ฆฌ์žฅ์น˜๊ฐ€ ์žˆ์„๊นŒ์š”?
02:36
And we invented the flying phone.
44
156080
2656
๊ทธ๋ž˜์„œ ์ €ํฌ๊ฐ€ ๋‚ ์•„๋‹ค๋‹ˆ๋Š” ์ „ํ™”๋ฅผ ๋ฐœ๋ช…ํ–ˆ์Šต๋‹ˆ๋‹ค.
02:38
(Laughter)
45
158760
1936
(์›ƒ์Œ)
02:40
So this robot uses a Samsung Galaxy smartphone that you can buy off the shelf,
46
160720
6176
์ด ๋กœ๋ด‡์€ ๋งค์žฅ์—์„œ ๊ตฌ์ž…ํ•  ์ˆ˜ ์žˆ๋Š” ์‚ผ์„ฑ ๊ฐค๋Ÿญ์‹œ ์Šค๋งˆํŠธํฐ์„ ์ผ๊ณ 
02:46
and all you need is an app that you can download from our app store.
47
166920
4016
์•ฑ์Šคํ† ์–ด์—์„œ ์•ฑ์„ ๋‹ค์šด๋กœ๋“œ ํ•˜๊ธฐ๋งŒ ํ•˜๋ฉด ๋ฉ๋‹ˆ๋‹ค.
02:50
And you can see this robot reading the letters, "TED" in this case,
48
170960
4216
๋กœ๋ด‡์ด ๊ธ€์ž๋„ ์ธ์‹ํ•ฉ๋‹ˆ๋‹ค. ์—ฌ๊ธฐ์„œ๋Š” "TED"๋ฅผ ์ฝ์ฃ .
02:55
looking at the corners of the "T" and the "E"
49
175200
2936
T์™€ E์˜ ๊ฐ€์žฅ์ž๋ฆฌ๋ฅผ ๋ณด๋ฉฐ
02:58
and then triangulating off of that, flying autonomously.
50
178160
3480
์ž๋™๋น„ํ–‰์„ ํ•˜๋ฉฐ ์‚ผ๊ฐ์ธก๋Ÿ‰์„ ํ•ฉ๋‹ˆ๋‹ค.
03:02
That joystick is just there to make sure if the robot goes crazy,
51
182720
3256
์กฐ์ด์Šคํ‹ฑ์ด ๋กœ๋ด‡์ด ๋งˆ๊ตฌ ์›€์ง์ด์ง€ ์•Š๊ฒŒ ์กฐ์ •ํ•ฉ๋‹ˆ๋‹ค.
์ฅฌ์„ธํŽ˜๊ฐ€ ๋กœ๋ด‡์„ ๋Œ ์ˆ˜ ์žˆ์ฃ .
03:06
Giuseppe can kill it.
52
186000
1416
03:07
(Laughter)
53
187440
1640
(์›ƒ์Œ)
03:10
In addition to building these small robots,
54
190920
3816
์ด๋Ÿฌํ•œ ์ž‘์€ ๋กœ๋ด‡์„ ์ œ์ž‘ํ•จ๊ณผ ๋”๋ถˆ์–ด
03:14
we also experiment with aggressive behaviors, like you see here.
55
194760
4800
์—ฌ๊ธฐ์„œ ๋ณด์‹œ๋Š” ์ด๋Ÿฐ ์—ญ๋™์ ์ธ ํ–‰๋™๋„ ์‹คํ—˜ํ•ฉ๋‹ˆ๋‹ค.
03:19
So this robot is now traveling at two to three meters per second,
56
199920
5296
์ด ๋กœ๋ด‡์€ ์ดˆ๋‹น 2-3m๋กœ ์›€์ง์ด๋ฉฐ
03:25
pitching and rolling aggressively as it changes direction.
57
205240
3496
๋ฐฉํ–ฅ์„ ๋ฐ”๊ฟ€ ๋•Œ ๊ณต๊ฒฉ์ ์œผ๋กœ ์˜ฌ๋ผ๊ฐ€๊ฑฐ๋‚˜ ๋•๋‹ˆ๋‹ค.
03:28
The main point is we can have smaller robots that can go faster
58
208760
4256
์š”์ ์€ ๋” ๋นจ๋ฆฌ ๊ฐˆ ์ˆ˜ ์žˆ๊ณ  ๋น„๊ตฌ์กฐํ™”๋œ ํ™˜๊ฒฝ์—์„œ ์›€์ง์ผ ์ˆ˜ ์žˆ๋Š”
03:33
and then travel in these very unstructured environments.
59
213040
2960
๋” ์ž‘์€ ๋กœ๋ด‡์„ ๊ฐ€์งˆ ์ˆ˜ ์žˆ๋‹ค๋Š” ๊ฒ๋‹ˆ๋‹ค.
03:37
And in this next video,
60
217120
2056
๋‹ค์Œ ์˜์ƒ์—์„œ๋Š”
03:39
just like you see this bird, an eagle, gracefully coordinating its wings,
61
219200
5896
์ƒˆ๋‚˜ ๋…์ˆ˜๋ฆฌ๊ฐ€ ์šฐ์•„ํ•˜๊ฒŒ ๋‚ ๊ฐœ์™€
03:45
its eyes and feet to grab prey out of the water,
62
225120
4296
๋ˆˆ, ๋‹ค๋ฆฌ๋ฅผ ์กฐ์ •ํ•ด์„œ ๋ฌผ ๋ฐ–์œผ๋กœ ๋จน์ด๋ฅผ ์žก์•„ ์ฑ„๋Š” ๊ฑธ ๋ณด์‹œ๋“ฏ์ด
03:49
our robot can go fishing, too.
63
229440
1896
์šฐ๋ฆฌ ๋กœ๋ด‡๋„ ๋‚š์‹œ๋ฅผ ํ•  ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค.
03:51
(Laughter)
64
231360
1496
(์›ƒ์Œ)
03:52
In this case, this is a Philly cheesesteak hoagie that it's grabbing out of thin air.
65
232880
4056
์ด ๊ฒฝ์šฐ์—๋Š” ๊ฐ‘์ž๊ธฐ ํ•„๋ฆฌ ์น˜์ฆˆ์Šคํ…Œ์ดํฌ๋ฅผ ๋‚š์•„ ์ฑ•๋‹ˆ๋‹ค.
03:56
(Laughter)
66
236960
2400
(์›ƒ์Œ)
03:59
So you can see this robot going at about three meters per second,
67
239680
3296
์ด ๋กœ๋ด‡์ด ์ดˆ๋‹น 3m๋กœ ์›€์ง์ด๋Š” ๊ฑธ ๋ณด์‹œ๋Š”๋ฐ
๊ฑท๋Š” ์†๋„๋ณด๋‹ค ๋น ๋ฅด๋ฉฐ ํŒ”, ๋ฐœํ†ฑ๊ณผ ๋น„ํ–‰์„
04:03
which is faster than walking speed, coordinating its arms, its claws
68
243000
5136
04:08
and its flight with split-second timing to achieve this maneuver.
69
248160
4120
๋ˆˆ ๊นœ์งํ•  ์‚ฌ์ด์— ์กฐ์ •ํ•ด์„œ ์ด๋ ‡๊ฒŒ ํ•˜๋Š” ๊ฒ๋‹ˆ๋‹ค.
04:14
In another experiment,
70
254120
1216
๋‹ค๋ฅธ ์‹คํ—˜์—์„œ๋Š”
04:15
I want to show you how the robot adapts its flight
71
255360
3656
๋กœ๋ด‡์ด ๋น„ํ–‰์„ ์–ด๋–ป๊ฒŒ ์กฐ์ ˆํ•˜์—ฌ
04:19
to control its suspended payload,
72
259040
2376
๋ถ€์œ  ํ•˜์ค‘์„ ํ†ต์ œํ•˜๋Š” ์ง€ ๋ณด์—ฌ๋“œ๋ฆฌ๊ฒ ์Šต๋‹ˆ๋‹ค.
04:21
whose length is actually larger than the width of the window.
73
261440
3800
์ฐฝ ๋„“์ด๋ณด๋‹ค ์‹ค์ œ๋กœ ๋” ํฐ ๊ธธ์ด์ž…๋‹ˆ๋‹ค.
04:25
So in order to accomplish this,
74
265680
1696
์ด๋ ‡๊ฒŒ ํ•˜๊ธฐ ์œ„ํ•ด์„œ
04:27
it actually has to pitch and adjust the altitude
75
267400
3696
๊ณ ๋„๋ฅผ ๋†’์ด๊ณ  ์กฐ์ ˆํ•˜์—ฌ
04:31
and swing the payload through.
76
271120
2320
์ ์žฌ ํ•˜์ค‘์„ ํ”๋“ค์–ด ํ†ต๊ณผํ•ฉ๋‹ˆ๋‹ค.
04:38
But of course we want to make these even smaller,
77
278920
2296
๋ฌผ๋ก  ์ด๊ฒƒ์„ ๋”์šฑ ์ž‘๊ฒŒ ํ•˜๊ณ  ์‹ถ์–ด์„œ
04:41
and we're inspired in particular by honeybees.
78
281240
3016
ํŠนํžˆ ๊ฟ€๋ฒŒ์—์„œ ์˜๊ฐ์„ ๋ฐ›์•˜์Šต๋‹ˆ๋‹ค.
04:44
So if you look at honeybees, and this is a slowed down video,
79
284280
3256
์ด๊ฒƒ์€ ๋Š๋ฆฐ ์˜์ƒ์ธ๋ฐ, ๊ฟ€๋ฒŒ์„ ๋ณด์‹œ๋ฉด
04:47
they're so small, the inertia is so lightweight --
80
287560
3720
์•„์ฃผ ์ž‘์€๋ฐ ๊ด€์„ฑ์ด ๋งค์šฐ ๊ฐ€๋ฒผ์›Œ์„œ
04:51
(Laughter)
81
291960
1176
(์›ƒ์Œ)
04:53
that they don't care -- they bounce off my hand, for example.
82
293160
3536
๊ฐ€๋ น ์ œ ์†์— ํŠ•๊ฒจ๋„ ์‹ ๊ฒฝ๋„ ์•ˆ ์”๋‹ˆ๋‹ค.
04:56
This is a little robot that mimics the honeybee behavior.
83
296720
3160
์ด๊ฒƒ์€ ๊ฟ€๋ฒŒ์˜ ํ–‰๋™์„ ํ‰๋‚ด๋‚ด๋Š” ์†Œํ˜• ๋กœ๋ด‡์ž…๋‹ˆ๋‹ค.
05:00
And smaller is better,
84
300600
1216
์ž‘์„์ˆ˜๋ก ๋” ์ข‹์€ ๊ฒƒ์ด
05:01
because along with the small size you get lower inertia.
85
301840
3536
์ด๋ ‡๊ฒŒ ์†Œ๊ทœ๋ชจ๋กœ ํ•˜๋ฉด ๊ด€์„ฑ์ด ๋” ๋‚ฎ์•„์ง‘๋‹ˆ๋‹ค.
05:05
Along with lower inertia --
86
305400
1536
๊ด€์„ฑ์ด ๋‚ฎ์•„์ง€๋ฉด --
05:06
(Robot buzzing, laughter)
87
306960
2856
(์œ™์œ™๊ฑฐ๋ฆฌ๋Š” ๋กœ๋ด‡) (์›ƒ์Œ)
05:09
along with lower inertia, you're resistant to collisions.
88
309840
2816
๊ด€์„ฑ์ด ๋‚ฎ์•„์ง€๋ฉด, ์ถฉ๋Œ์— ์ €ํ•ญ๋ ฅ์ด ์žˆ์Šต๋‹ˆ๋‹ค.
05:12
And that makes you more robust.
89
312680
1720
๊ทธ๊ฒƒ์ด ๋”์šฑ ํŠผํŠผํ•˜๊ฒŒ ํ•ด ์ค๋‹ˆ๋‹ค.
05:15
So just like these honeybees, we build small robots.
90
315800
2656
์ €ํฌ๋Š” ๊ฟ€๋ฒŒ๊ฐ™์€ ์†Œํ˜• ๋กœ๋ด‡์„ ๋งŒ๋“ญ๋‹ˆ๋‹ค.
05:18
And this particular one is only 25 grams in weight.
91
318480
3376
ํŠนํžˆ ์ด๊ฒƒ์€ 25g๋ฐ–์— ์•ˆ ๋ฉ๋‹ˆ๋‹ค.
05:21
It consumes only six watts of power.
92
321880
2160
์ „๋ ฅ๋„ 6์™€ํŠธ๋งŒ ์†Œ๋ชจํ•˜์ง€์š”.
05:24
And it can travel up to six meters per second.
93
324440
2536
์ดˆ๋‹น 6m๊นŒ์ง€ ์›€์ง์ผ ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค.
์ œ๊ฐ€ ์ด๊ฒƒ์„ ์ด ํฌ๊ธฐ์— ํ‘œ์ค€ํ™”์‹œํ‚ค๋ฉด
05:27
So if I normalize that to its size,
94
327000
2336
05:29
it's like a Boeing 787 traveling ten times the speed of sound.
95
329360
3640
๋ณด์ž‰787์ด ์Œ์†์˜ 10๋ฐฐ๋กœ ๋‚˜๋Š” ๊ฒƒ๊ณผ ๊ฐ™์Šต๋‹ˆ๋‹ค.
05:36
(Laughter)
96
336000
2096
(์›ƒ์Œ)
05:38
And I want to show you an example.
97
338120
1920
์ œ๊ฐ€ ์˜ˆ์‹œ๋ฅผ ๋ณด์—ฌ๋“œ๋ฆฌ๊ฒ ์Šต๋‹ˆ๋‹ค.
05:40
This is probably the first planned mid-air collision, at one-twentieth normal speed.
98
340840
5256
์ด๊ฒƒ์ด ์ตœ์ดˆ์˜ ๊ณ„ํš๋œ ๊ณต์ค‘ ์ถฉ๋Œ์ผํ…๋ฐ์š”, 1/20 ํ‘œ์ค€์†์ž…๋‹ˆ๋‹ค.
05:46
These are going at a relative speed of two meters per second,
99
346120
2858
์ดˆ๋‹น 2m์˜ ์ƒ๋Œ€์†์œผ๋กœ ๊ฐ‘๋‹ˆ๋‹ค.
์ด๊ฒƒ์ด ๊ธฐ๋ณธ ์›๋ฆฌ๋ฅผ ์ž˜ ์„ค๋ช…ํ•ด ์ค๋‹ˆ๋‹ค.
05:49
and this illustrates the basic principle.
100
349002
2480
05:52
The two-gram carbon fiber cage around it prevents the propellers from entangling,
101
352200
4976
2g์งœ๋ฆฌ ํƒ„์†Œ์„ฌ์œ ๋กœ ๋‘๋ฅด๋ฉด ํ”„๋กœํŽ ๋Ÿฌ๊ฐ€ ์—‰ํ‚ค๋Š” ๊ฑธ ๋ฐฉ์ง€ํ•˜์ง€๋งŒ
05:57
but essentially the collision is absorbed and the robot responds to the collisions.
102
357200
5296
๊ธฐ๋ณธ์ ์œผ๋กœ ์ถฉ๋Œ์ด ํก์ˆ˜๋˜๊ณ  ๋กœ๋ด‡์ด ์ถฉ๋Œ์— ๋ฐ˜์‘ํ•ฉ๋‹ˆ๋‹ค.
06:02
And so small also means safe.
103
362520
2560
๋งค์šฐ ์ž‘๋‹ค๋Š” ๊ฑด ๋˜ํ•œ ์•ˆ์ „ํ•˜๋‹ค๋Š” ๋œป์ž…๋‹ˆ๋‹ค.
06:05
In my lab, as we developed these robots,
104
365400
2016
์ œ ์—ฐ๊ตฌ์‹ค์—์„œ ์ด ๋กœ๋ด‡์„ ๊ฐœ๋ฐœํ–ˆ์„ ๋•Œ
06:07
we start off with these big robots
105
367440
1620
์ด๋Ÿฐ ๋Œ€ํ˜• ๋กœ๋ด‡๋„ ์‹œ์ž‘ํ–ˆ๊ณ 
06:09
and then now we're down to these small robots.
106
369084
2812
์ง€๊ธˆ์€ ์ด๋Ÿฐ ์†Œํ˜• ๋กœ๋ด‡์— ๋งค์ง„ํ•ฉ๋‹ˆ๋‹ค.
06:11
And if you plot a histogram of the number of Band-Aids we've ordered
107
371920
3456
์ €ํฌ๊ฐ€ ๊ณผ๊ฑฐ์— ์ฃผ๋ฌธํ•œ ๋ฐ˜์ฐฝ๊ณ  ์ˆ˜๋ฅผ ๊ทธ๋ž˜ํ”„๋กœ ๊ทธ๋ฆฌ๋ฉด
06:15
in the past, that sort of tailed off now.
108
375400
2576
์ง€๊ธˆ์€ ๊ทธ๋ž˜ํ”„๊ฐ€ ์ฐจ์ฐจ ๊ฐ์†Œํ–ˆ์„ ๊ฒ๋‹ˆ๋‹ค.
์ด ๋กœ๋ด‡์ด ๋งค์šฐ ์•ˆ์ „ํ•˜๊ธฐ ๋•Œ๋ฌธ์ž…๋‹ˆ๋‹ค.
06:18
Because these robots are really safe.
109
378000
1960
06:20
The small size has some disadvantages,
110
380760
2456
์†Œํ˜• ํฌ๊ธฐ๋Š” ๋‹จ์ ์ด ์žˆ์Šต๋‹ˆ๋‹ค.
06:23
and nature has found a number of ways to compensate for these disadvantages.
111
383240
4080
์ž์—ฐ์€ ์ด๋Ÿฐ ๋‹จ์ ๋“ค์„ ์—ฌ๋Ÿฌ๊ฐ€์ง€ ๋ฐฉ๋ฒ•์œผ๋กœ ๋ณด์ƒํ•  ๋ฐฉ๋ฒ•์„ ์ฐพ์•˜์ฃ .
06:27
The basic idea is they aggregate to form large groups, or swarms.
112
387960
4000
๊ธฐ๋ณธ์ ์ธ ์›๋ฆฌ๋Š” ํฐ ์ง‘๋‹จ, ๋˜๋Š” ๋–ผ๋ฅผ ์ง“๋Š” ๊ฒ๋‹ˆ๋‹ค.
06:32
So, similarly, in our lab, we try to create artificial robot swarms.
113
392320
3976
์—ฐ๊ตฌ์‹ค์—์„œ ๋น„์Šทํ•˜๊ฒŒ ์ธ๊ณต์ ์ธ ๋กœ๋ด‡ ๋–ผ๋ฅผ ๋งŒ๋“ค์–ด ๋ดค์Šต๋‹ˆ๋‹ค.
06:36
And this is quite challenging
114
396320
1381
์ด๊ฑด ๋งค์šฐ ํž˜๋“ค์—ˆ๋Š”๋ฐ
06:37
because now you have to think about networks of robots.
115
397725
3320
๋กœ๋ด‡์˜ ์—ฐ๊ฒฐ๋ง์„ ์ƒ๊ฐํ•ด์•ผ ํ–ˆ๊ธฐ ๋•Œ๋ฌธ์ž…๋‹ˆ๋‹ค.
06:41
And within each robot,
116
401360
1296
๊ฐ๊ฐ์˜ ๋กœ๋ด‡ ์•ˆ์—์„œ
06:42
you have to think about the interplay of sensing, communication, computation --
117
402680
5616
๊ฐ์ง€, ์˜์‚ฌ์†Œํ†ต, ์‹ ํ˜ธ์ฒ˜๋ฆฌ์˜ ์ƒํ˜ธ์ž‘์šฉ์„ ์ƒ๊ฐํ•ด์•ผ ํ•˜๋Š”๋ฐ
06:48
and this network then becomes quite difficult to control and manage.
118
408320
4960
์ด ์—ฐ๊ฒฐ๋ง์€ ํ†ต์ œ ๊ด€๋ฆฌ๊ฐ€ ๋งค์šฐ ์–ด๋ ต๊ฒŒ ๋ฉ๋‹ˆ๋‹ค.
06:54
So from nature we take away three organizing principles
119
414160
3296
๊ทธ๋ž˜์„œ ์ž์—ฐ์—์„œ ํ•„์ˆ˜์ ์œผ๋กœ ์•Œ๊ณ ๋ฆฌ์ฆ˜์„ ์งœ๊ฒŒ ํ•œ ์„ธ ๊ฐ€์ง€ ์กฐ์ง ์›๋ฆฌ๋ฅผ ๋ฝ‘์•˜์Šต๋‹ˆ๋‹ค.
06:57
that essentially allow us to develop our algorithms.
120
417480
3160
07:01
The first idea is that robots need to be aware of their neighbors.
121
421640
4536
์ฒซ ๋ฒˆ์งธ ์›๋ฆฌ๋Š” ๋กœ๋ด‡์ด ๊ทผ์ ‘ํ•œ ๋กœ๋ด‡์„ ์ธ์ง€ํ•ด์•ผ ํ•˜๋Š” ๊ฒ๋‹ˆ๋‹ค.
07:06
They need to be able to sense and communicate with their neighbors.
122
426200
3440
๋‹ค๋ฅธ ๋กœ๋ด‡๋“ค์„ ์ธ์‹ํ•˜๊ณ  ์†Œํ†ตํ•  ์ˆ˜ ์žˆ์–ด์•ผ ํ•ฉ๋‹ˆ๋‹ค.
07:10
So this video illustrates the basic idea.
123
430040
2656
์ด ์˜์ƒ์€ ๊ทธ ๊ธฐ๋ณธ ์›๋ฆฌ๋ฅผ ๋ณด์—ฌ์ค๋‹ˆ๋‹ค.
07:12
You have four robots --
124
432720
1296
๋กœ๋ด‡์ด ๋„ค ๋Œ€๊ฐ€ ์žˆ์Šต๋‹ˆ๋‹ค.
ํ•œ ๋กœ๋ด‡์ด ์ธ๊ฐ„ ์กฐ์ข…์‚ฌ์—๊ฒŒ ๊ทธ์•ผ ๋ง๋กœ ๋‚ฉ์น˜๋˜์—ˆ์Šต๋‹ˆ๋‹ค.
07:14
one of the robots has actually been hijacked by a human operator, literally.
125
434040
4240
07:19
But because the robots interact with each other,
126
439217
2239
๊ทธ๋Ÿฌ๋‚˜ ๋กœ๋ด‡์ด ์„œ๋กœ ์ƒํ˜ธ์ž‘์šฉํ•˜๊ธฐ ๋•Œ๋ฌธ์—
07:21
they sense their neighbors,
127
441480
1656
๋‹ค๋ฅธ ๋กœ๋ด‡์„ ๊ฐ์ง€ํ•˜๊ณ  ๊ธฐ๋ณธ์ ์œผ๋กœ ๋”ฐ๋ผ๊ฐ‘๋‹ˆ๋‹ค.
07:23
they essentially follow.
128
443160
1296
07:24
And here there's a single person able to lead this network of followers.
129
444480
5360
์—ฌ๊ธฐ ํ•œ ๋ช…์ด ์ด ์ถ”์ข…๋กœ๋ด‡์˜ ์—ฐ๊ฒฐ๋ง์„ ์ฃผ๋„ํ•ฉ๋‹ˆ๋‹ค.
07:32
So again, it's not because all the robots know where they're supposed to go.
130
452000
5056
๋‹ค์‹œ ๋งํ•˜๋ฉด ๋ชจ๋“  ๋กœ๋ด‡์ด ์–ด๋””๋กœ ๊ฐˆ์ง€ ์•Œ๊ธฐ ๋•Œ๋ฌธ์ด ์•„๋‹ˆ๋ผ
07:37
It's because they're just reacting to the positions of their neighbors.
131
457080
4320
๊ทผ์ ‘ ๋กœ๋ด‡๋“ค์˜ ์œ„์น˜์— ๊ทธ์ € ๋ฐ˜์‘ํ•˜๊ธฐ ๋•Œ๋ฌธ์ธ ๊ฒ๋‹ˆ๋‹ค.
07:43
(Laughter)
132
463720
4120
(์›ƒ์Œ)
07:48
So the next experiment illustrates the second organizing principle.
133
468280
5240
๋‹ค์Œ ์‹คํ—˜์€ ๋‘ ๋ฒˆ์งธ ์กฐ์ง ์›๋ฆฌ๋ฅผ ๋ณด์—ฌ์ค๋‹ˆ๋‹ค.
07:54
And this principle has to do with the principle of anonymity.
134
474920
3800
์ด ์›๋ฆฌ๋Š” ์ต๋ช…์˜ ์›๋ฆฌ์™€ ๊ด€๊ณ„๊ฐ€ ์žˆ์Šต๋‹ˆ๋‹ค.
07:59
Here the key idea is that
135
479400
4296
์ค‘์‹ฌ ์ƒ๊ฐ์€
08:03
the robots are agnostic to the identities of their neighbors.
136
483720
4240
๋กœ๋ด‡๋“ค์ด ๋‹ค๋ฅธ ๋กœ๋ด‡์˜ ์ •์ฒด๋ฅผ ๋ชจ๋ฅธ๋‹ค๋Š” ๊ฒƒ์ด์ฃ .
08:08
They're asked to form a circular shape,
137
488440
2616
์ด๋“ค์ด ์›ํ˜•์„ ๋งŒ๋“ค๋„๋ก ๋ช…๋ น ๋ฐ›์œผ๋ฉด
08:11
and no matter how many robots you introduce into the formation,
138
491080
3296
์ด ๊ณผ์ •์— ๋ช‡ ๊ฐœ์˜ ๋กœ๋ด‡์„ ํˆฌ์ž…ํ•˜๊ฑฐ๋‚˜
08:14
or how many robots you pull out,
139
494400
2576
๋ช‡ ๊ฐœ๋ฅผ ๋นผ๋“ ์ง€
๊ฐ ๋กœ๋ด‡์€ ๋‹จ์ง€ ๋‹ค๋ฅธ ๋กœ๋ด‡์— ๋ฐ˜์‘ํ•ฉ๋‹ˆ๋‹ค.
08:17
each robot is simply reacting to its neighbor.
140
497000
3136
08:20
It's aware of the fact that it needs to form the circular shape,
141
500160
4976
์›ํ˜•์„ ๋งŒ๋“ค์–ด์•ผ ํ•˜๋Š” ์‚ฌ์‹ค์„ ์ธ์ง€ํ•˜์ง€๋งŒ
08:25
but collaborating with its neighbors
142
505160
1776
๋‹ค๋ฅธ ๋กœ๋ด‡๊ณผ ํ˜‘๋ ฅํ•˜์—ฌ
08:26
it forms the shape without central coordination.
143
506960
3720
์ค‘์•™ ํ†ต์ œ๊ฐ€ ์—†์ด ๋ชจ์–‘์„ ํ˜•์„ฑํ•ฉ๋‹ˆ๋‹ค.
08:31
Now if you put these ideas together,
144
511520
2416
์ด ์ƒ๊ฐ๋“ค์„ ํ•œ๋ฐ ๋ชจ์œผ๋ฉด
08:33
the third idea is that we essentially give these robots
145
513960
3896
์„ธ ๋ฒˆ์งธ ์ƒ๊ฐ์ด ๊ธฐ๋ณธ์ ์œผ๋กœ ์ €ํฌ๋Š” ์ด ๋กœ๋ด‡๋“ค์—๊ฒŒ
08:37
mathematical descriptions of the shape they need to execute.
146
517880
4296
์‹คํ–‰ํ•ด์•ผ ํ•  ๋ชจ์–‘์˜ ์ˆ˜ํ•™์  ์„ค๋ช…์„ ์ฃผ๋Š” ๊ฒ๋‹ˆ๋‹ค.
08:42
And these shapes can be varying as a function of time,
147
522200
3496
์ด๋Ÿฐ ๋ชจ์–‘์€ ์‹œ๊ฐ„์˜ ํ•จ์ˆ˜์— ๋”ฐ๋ผ ๋‹ค๋ฅผ ์ˆ˜ ์žˆ๋Š”๋ฐ
08:45
and you'll see these robots start from a circular formation,
148
525720
4496
์ด ๋กœ๋ด‡๋“ค์ด ์›ํ˜•์œผ๋กœ ์›€์ง์ด๊ธฐ ์‹œ์ž‘ํ•˜๊ณ 
08:50
change into a rectangular formation, stretch into a straight line,
149
530240
3256
์ง์‚ฌ๊ฐํ˜• ๋ชจ์–‘์œผ๋กœ ๋ฐ”๋€Œ๊ณ  ์ง์„ ์œผ๋กœ ์ญ‰ ๋ป—์–ด
08:53
back into an ellipse.
150
533520
1375
ํƒ€์›ํ˜•์œผ๋กœ ๋‹ค์‹œ ๋Œ์•„๊ฐ€๋Š” ๊ฒŒ ๋ณด์ด์‹ค ๊ฒ๋‹ˆ๋‹ค.
08:54
And they do this with the same kind of split-second coordination
151
534919
3617
๋กœ๋ด‡๋“ค์ด ์ˆœ๊ฐ„์ ์ธ ์กฐ์ •์œผ๋กœ ํ•˜๋Š” ๊ฒƒ์ธ๋ฐ
08:58
that you see in natural swarms, in nature.
152
538560
3280
์ž์—ฐ์—์„œ ๋ณด๋Š” ๋ฒŒ๋–ผ์™€ ๊ฐ™์€ ๊ฒƒ์ด์ฃ .
09:03
So why work with swarms?
153
543080
2136
์™œ ๋–ผ๋กœ ์ž‘์—…ํ•˜๋Š๋ƒ๊ณ ์š”?
09:05
Let me tell you about two applications that we are very interested in.
154
545240
4120
์ €ํฌ๊ฐ€ ๋งค์šฐ ํฅ๋ฏธ๋ฅผ ๊ฐ–๊ณ  ์žˆ๋Š” ๋‘ ๊ฐ€์ง€ ์‘์šฉ๋ฐฉ๋ฒ•์„ ๋ง์”€๋“œ๋ฆฌ๊ฒ ์Šต๋‹ˆ๋‹ค.
09:10
The first one has to do with agriculture,
155
550160
2376
์ฒซ ๋ฒˆ์งธ๋Š” ๋†์—…๊ณผ ๊ด€๋ จ๋˜์–ด ์žˆ๋Š”๋ฐ
09:12
which is probably the biggest problem that we're facing worldwide.
156
552560
3360
์ „ ์„ธ๊ณ„์ ์œผ๋กœ ์šฐ๋ฆฌ๊ฐ€ ๋‹น๋ฉดํ•œ ์ตœ๋Œ€์˜ ๋ฌธ์ œ์ผ ๊ฒ๋‹ˆ๋‹ค.
09:16
As you well know,
157
556760
1256
์—ฌ๋Ÿฌ๋ถ„์ด ์ž˜ ์•„์‹œ๋“ฏ์ด
09:18
one in every seven persons in this earth is malnourished.
158
558040
3520
์ง€๊ตฌ์ƒ์— 7๋ช… ์ค‘ ํ•œ ๋ช…์ด ์˜์–‘์‹ค์กฐ์ž…๋‹ˆ๋‹ค.
09:21
Most of the land that we can cultivate has already been cultivated.
159
561920
3480
๊ฒฝ์ž‘ํ•  ์ˆ˜ ์žˆ๋Š” ๋•…์€ ๋‹ค ๋†์‚ฌ์ง“๊ณ  ์žˆ์Šต๋‹ˆ๋‹ค.
09:25
And the efficiency of most systems in the world is improving,
160
565960
3216
์„ธ๊ณ„์˜ ๋Œ€๋ถ€๋ถ„ ์‹œ์Šคํ…œ์˜ ํšจ์œจ์„ฑ์ด ํ–ฅ์ƒ๋˜๊ณ  ์žˆ์ง€๋งŒ
09:29
but our production system efficiency is actually declining.
161
569200
3520
์ƒ์‚ฐ ์‹œ์Šคํ…œ์˜ ํšจ์œจ์„ฑ์€ ์‹ค์ œ๋กœ ๊ฐ์†Œํ•˜๊ณ  ์žˆ์Šต๋‹ˆ๋‹ค.
09:33
And that's mostly because of water shortage, crop diseases, climate change
162
573080
4216
๋Œ€๋ถ€๋ถ„ ๋ฌผ ๋ถ€์กฑ, ์ž‘๋ฌผ ๋ณ‘, ๊ธฐํ›„ ๋ณ€ํ™”์™€
09:37
and a couple of other things.
163
577320
1520
๋‹ค๋ฅธ ๋ช‡ ๊ฐ€์ง€๋“ค ๋•Œ๋ฌธ์ž…๋‹ˆ๋‹ค.
09:39
So what can robots do?
164
579360
1480
๋กœ๋ด‡์ด ๋ฌด์—‡์„ ํ•  ์ˆ˜ ์žˆ์„๊นŒ์š”?
09:41
Well, we adopt an approach that's called Precision Farming in the community.
165
581200
4616
๋†์—…๊ณ„์—๋Š” ์ •๋ฐ€ ๋†๋ฒ•์ด๋ผ๊ณ  ํ•˜๋Š” ๋ฐฉ๋ฒ•์„ ์ฑ„ํƒํ•ฉ๋‹ˆ๋‹ค.
09:45
And the basic idea is that we fly aerial robots through orchards,
166
585840
5376
๊ธฐ๋ณธ ์›๋ฆฌ๋Š” ๊ณผ์ˆ˜์›์— ๋กœ๋ด‡์„ ๋น„ํ–‰์‹œ์ผœ์„œ
09:51
and then we build precision models of individual plants.
167
591240
3120
๊ฐ ์ž‘๋ฌผ์˜ ์ •๋ฐ€ํ•œ ๋ชจํ˜•์„ ๊ตฌ์ถ•ํ•˜๋Š” ๊ฒ๋‹ˆ๋‹ค.
09:54
So just like personalized medicine,
168
594829
1667
๋งˆ์น˜ ๊ฐœ์ธ๋ณ„ ๋งž์ถค ์˜ํ•™์ฒ˜๋Ÿผ
09:56
while you might imagine wanting to treat every patient individually,
169
596520
4816
๊ฐœ๋ณ„ ํ™˜์ž๋“ค์„ ๋ชจ๋‘ ๋‹ค๋ฅด๊ฒŒ ์น˜๋ฃŒํ•˜๋Š” ๊ฒƒ์„ ์ƒ์ƒํ•˜์‹œ๋Š” ๊ฒƒ ์ฒ˜๋Ÿผ
10:01
what we'd like to do is build models of individual plants
170
601360
3696
์ €ํฌ๊ฐ€ ํ•˜๊ณ  ์‹ถ์€ ๊ฒƒ์ด ๊ฐœ๋ณ„ ์ž‘๋ฌผ์˜ ๋ชจํ˜•์„ ๋งŒ๋“ค์–ด์„œ
10:05
and then tell the farmer what kind of inputs every plant needs --
171
605080
4136
๋†๋ถ€์—๊ฒŒ ๊ฐ ์ž‘๋ฌผ์ด ํ•„์š”ํ•œ ์‚ฌํ•ญ์„ ์•Œ๋ ค์ฃผ๋Š” ๊ฒ๋‹ˆ๋‹ค.
10:09
the inputs in this case being water, fertilizer and pesticide.
172
609240
4440
์ด ๊ฒฝ์šฐ์—๋Š” ๋ฌผ, ๋น„๋ฃŒ, ๋†์•ฝ์„ ์ฃผ๋Š” ๊ฒƒ์ด์ฃ .
10:14
Here you'll see robots traveling through an apple orchard,
173
614640
3616
์—ฌ๊ธฐ ๋ณด์‹œ๋ฉด ์‚ฌ๊ณผ ๋†์žฅ์„ ๋น„ํ–‰ํ•˜๋Š” ๋กœ๋ด‡์ด ๋ณด์ด๋Š”๋ฐ
10:18
and in a minute you'll see two of its companions
174
618280
2256
์ž ์‹œ ํ›„์— ๋‘ ๋Œ€์˜ ๋™๋ฃŒ ๋กœ๋ด‡์ด
10:20
doing the same thing on the left side.
175
620560
1810
์™ผ์ชฝ์—์„œ ๊ฐ™์€ ์ผ์„ ํ•˜๊ณ  ์žˆ๋Š”๊ฒŒ ๋ณด์ผ ๊ฒ๋‹ˆ๋‹ค.
10:22
And what they're doing is essentially building a map of the orchard.
176
622800
3656
์ด ๋กœ๋ด‡์ด ํ•˜๋Š” ์ผ์€ ๊ธฐ๋ณธ์ ์œผ๋กœ ๊ณผ์ˆ˜์›์˜ ์ง€๋„๋ฅผ ์ œ์ž‘ํ•˜๋Š” ๊ฒ๋‹ˆ๋‹ค.
10:26
Within the map is a map of every plant in this orchard.
177
626480
2816
์ง€๋„ ์•ˆ์—๋Š” ๊ณผ์ˆ˜์›์˜ ๋ชจ๋“  ์ž‘๋ถˆ์ด ๋“ค์–ด ๊ฐ‘๋‹ˆ๋‹ค.
10:29
(Robot buzzing)
178
629320
1656
(์œ™์œ™๊ฑฐ๋ฆฌ๋Š” ๋กœ๋ด‡)
๊ทธ ์ง€๋„๊ฐ€ ์–ด๋–ป๊ฒŒ ์ƒ๊ฒผ๋Š”์ง€ ๋ณด๊ฒ ์Šต๋‹ˆ๋‹ค.
10:31
Let's see what those maps look like.
179
631000
1896
10:32
In the next video, you'll see the cameras that are being used on this robot.
180
632920
4296
๋‹ค์Œ ์˜์ƒ์—์„œ ์ด ๋กœ๋ด‡์— ์“ฐ์ด๋Š” ์นด๋ฉ”๋ผ๊ฐ€ ๋ณด์ž…๋‹ˆ๋‹ค.
10:37
On the top-left is essentially a standard color camera.
181
637240
3240
๋งจ ์œ„ ์™ผ์ชฝ์— ๋›ฐ์–ด๋‚œ ์ปฌ๋Ÿฌ ์นด๋ฉ”๋ผ๊ฐ€ ์žˆ์Šต๋‹ˆ๋‹ค.
10:41
On the left-center is an infrared camera.
182
641640
3296
์™ผ์ชฝ ์ค‘์•™์—๋Š” ์ ์™ธ์„  ์นด๋ฉ”๋ผ๊ฐ€ ์žˆ์Šต๋‹ˆ๋‹ค.
10:44
And on the bottom-left is a thermal camera.
183
644960
3776
์•„๋ž˜ ์™ผ์ชฝ์—๋Š” ์˜จ๋„๊ฐ์ง€ ์นด๋ฉ”๋ผ๊ฐ€ ์žˆ์ฃ .
10:48
And on the main panel, you're seeing a three-dimensional reconstruction
184
648760
3336
์ฃผ ํŒจ๋„์—๋Š” 3์ฐจ์›์œผ๋กœ ๊ณผ์ˆ˜์›์˜ ๋ชจ๋“  ๋‚˜๋ฌด๊ฐ€ ์žฌ๊ตฌ์„ฑ๋œ ๊ฒƒ์„ ๋ณด๊ณ  ๊ณ„์‹ญ๋‹ˆ๋‹ค.
10:52
of every tree in the orchard as the sensors fly right past the trees.
185
652120
6120
์„ผ์„œ๊ฐ€ ๋‚˜๋ฌด ์‚ฌ์ด๋ฅผ ๋น„ํ–‰ํ•˜๋ฉด์„œ ์ƒ๊ธด ๊ฒƒ์ด์ฃ .
10:59
Armed with information like this, we can do several things.
186
659640
4040
์ด๋Ÿฐ ์ •๋ณด๋ฅผ ๊ฐ–์ถ”๊ณ  ์žˆ์œผ๋ฉด ๋ช‡ ๊ฐ€์ง€๋ฅผ ํ•  ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค.
11:04
The first and possibly the most important thing we can do is very simple:
187
664200
4256
๊ฐ€์žฅ ์ค‘์š”ํ•œ ์ฒซ ๋ฒˆ์งธ๋Š” ๋งค์šฐ ๋‹จ์ˆœํ•ฉ๋‹ˆ๋‹ค.
11:08
count the number of fruits on every tree.
188
668480
2440
๋ชจ๋“  ๋‚˜๋ฌด์˜ ์—ด๋งค๋ฅผ ์„ธ๋Š” ๊ฒ๋‹ˆ๋‹ค.
11:11
By doing this, you tell the farmer how many fruits she has in every tree
189
671520
4536
์ด๋ ‡๊ฒŒ ํ•˜๋ฉด ๋†๋ถ€์—๊ฒŒ ๋ชจ๋“  ๋‚˜๋ฌด์˜ ์—ด๋งค ์ˆ˜๋ฅผ ์•Œ๋ ค์ฃผ๊ณ 
11:16
and allow her to estimate the yield in the orchard,
190
676080
4256
๊ณผ์ˆ˜์›์˜ ์ˆ˜ํ™•๋Ÿ‰์„ ์ถ”์ •ํ•˜๊ฒŒ ํ•ด ์ค˜์„œ
11:20
optimizing the production chain downstream.
191
680360
2840
์ƒ์‚ฐ๊ณผ์ •์„ ์ตœ์ ํ™” ํ•˜๊ฒŒ ํ•ด ์ค๋‹ˆ๋‹ค.
11:23
The second thing we can do
192
683640
1616
๋‘ ๋ฒˆ์งธ๋กœ ํ•  ์ˆ˜ ์žˆ๋Š” ์ผ์€
11:25
is take models of plants, construct three-dimensional reconstructions,
193
685280
4496
์ž‘๋ฌผ์˜ ๋ชจํ˜•์„ ๋”ฐ์„œ 3์ฐจ์›์œผ๋กœ ์žฌ๊ตฌ์„ฑํ•˜๊ณ 
11:29
and from that estimate the canopy size,
194
689800
2536
๋ฎ๊ฐœ ํฌ๊ธฐ ์ถ”์ •์น˜๋กœ
11:32
and then correlate the canopy size to the amount of leaf area on every plant.
195
692360
3776
๋ชจ๋“  ์ž‘๋ฌผ์˜ ์žŽ ๊ฐœ์ˆ˜์™€ ๋ฎ๊ฐœ ํฌ๊ธฐ์˜ ์ƒ๊ด€๊ด€๊ณ„๋ฅผ ์•Œ์•„ ๋ณด๋Š” ๊ฒ๋‹ˆ๋‹ค.
11:36
And this is called the leaf area index.
196
696160
2176
์ด๊ฒƒ์€ ์žŽ์‚ฌ๊ท€ ์ง€ํ‘œ๋ผ๊ณ  ํ•ฉ๋‹ˆ๋‹ค.
11:38
So if you know this leaf area index,
197
698360
1936
์ด๋Ÿฐ ์žŽ์‚ฌ๊ท€ ์ง€ํ‘œ๋ฅผ ์•„์‹ ๋‹ค๋ฉด
11:40
you essentially have a measure of how much photosynthesis is possible in every plant,
198
700320
5456
๋ชจ๋“  ์ž‘๋ฌผ์˜ ๊ฐ€๋Šฅํ•œ ๊ด‘ํ•ฉ์„ฑ ์ˆ˜์น˜๋ฅผ ์ธก์ •ํ•˜๋Š” ๊ฒƒ์ด๊ณ 
11:45
which again tells you how healthy each plant is.
199
705800
2880
๊ทธ๋Ÿผ ๊ฐ ๋‚˜๋ฌด์˜ ๊ฑด๊ฐ•์ •๋„๋ฅผ ์•Œ ์ˆ˜ ์žˆ๋Š” ๊ฒ๋‹ˆ๋‹ค.
11:49
By combining visual and infrared information,
200
709520
4216
์‹œ๊ฐ ์ •๋ณด์™€ ์ ์™ธ์„  ์ •๋ณด๋ฅผ ๊ฒฐํ•ฉํ•˜๋ฉด
11:53
we can also compute indices such as NDVI.
201
713760
3296
NDVI๊ฐ™์€ ์ง€ํ‘œ๋ฅผ ๊ณ„์‚ฐํ•ด ๋‚ผ ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค.
11:57
And in this particular case, you can essentially see
202
717080
2816
ํŠนํžˆ ์ด ๊ฒฝ์šฐ์—๋Š” ์‹ค์ œ๋กœ
11:59
there are some crops that are not doing as well as other crops.
203
719920
3016
๋‹ค๋ฅธ ์ž‘๋ฌผ๋งŒํผ ์ข‹์ง€ ์•Š์€ ์ž‘๋ฌผ์ด ์žˆ์Œ์„ ์•Œ ์ˆ˜ ์žˆ์ฃ .
12:02
This is easily discernible from imagery,
204
722960
4056
์ด๋ฏธ์ง€๋ฅผ ํ†ตํ•ด ์‰ฝ๊ฒŒ ๊ตฌ๋ณ„ํ•  ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค.
๊ทธ์ € ์‹œ๊ฐ ์ด๋ฏธ์ง€๊ฐ€ ์•„๋‹ˆ๋ผ
12:07
not just visual imagery but combining
205
727040
2216
12:09
both visual imagery and infrared imagery.
206
729280
2776
์‹œ๊ฐ๊ณผ ์ ์™ธ์„  ์ด๋ฏธ์ง€๋ฅผ ๊ฒฐํ•ฉํ•˜๋Š” ๊ฒƒ์ด์ฃ .
12:12
And then lastly,
207
732080
1336
๋งˆ์ง€๋ง‰์œผ๋กœ
12:13
one thing we're interested in doing is detecting the early onset of chlorosis --
208
733440
4016
์ €ํฌ๊ฐ€ ํฅ๋ฏธ๋ฅผ ๊ฐ–๊ณ  ์žˆ๋Š” ๊ฒƒ์ด ๋ฐฑํ™” ์ดˆ๊ธฐ์ƒํƒœ๋ฅผ ํƒ์ง€ํ•˜๋Š” ๊ฒ๋‹ˆ๋‹ค.
12:17
and this is an orange tree --
209
737480
1496
์ด๊ฒƒ์€ ์˜ค๋ Œ์ง€ ๋‚˜๋ฌด์ธ๋ฐ
์žŽ์ด ํ™ฉ์ƒ‰์œผ๋กœ ๋ณด์ด์‹œ์ฃ .
12:19
which is essentially seen by yellowing of leaves.
210
739000
2560
12:21
But robots flying overhead can easily spot this autonomously
211
741880
3896
ํ•˜์ง€๋งŒ ๋กœ๋ด‡์ด ์œ„์—์„œ ๋น„ํ–‰ํ•˜๋ฉด ์‰ฝ๊ฒŒ ๋ฐœ๊ฒฌํ•˜๊ณ 
12:25
and then report to the farmer that he or she has a problem
212
745800
2936
๋†๋ถ€์—๊ฒŒ
12:28
in this section of the orchard.
213
748760
1520
๊ณผ์ˆ˜์›์˜ ์–ด๋Š ๋ถ€๋ถ„์— ๋ฌธ์ œ๊ฐ€ ์žˆ๋Š”์ง€ ๋ณด๊ณ ํ•ฉ๋‹ˆ๋‹ค.
12:30
Systems like this can really help,
214
750800
2696
์ด๋Ÿฐ ์‹œ์Šคํ…œ์€ ์ •๋ง ๋„์›€์ด ๋˜๋Š”๋ฐ
12:33
and we're projecting yields that can improve by about ten percent
215
753520
5816
์•ฝ 10% ์ •๋„ ์ˆ˜ํ™•๋Ÿ‰์„ ํ–ฅ์ƒ์‹œํ‚ฌ ๊ฒƒ์œผ๋กœ ์ „๋งํ•˜๊ณ  ์žˆ์Šต๋‹ˆ๋‹ค.
12:39
and, more importantly, decrease the amount of inputs such as water
216
759360
3216
๋”์šฑ ์ค‘์š”ํ•œ ๊ฒƒ์€ ๋ฌผ์„ ์ฃผ๋Š” ๊ฒƒ์„
12:42
by 25 percent by using aerial robot swarms.
217
762600
3280
๋กœ๋ด‡ ๋–ผ๋ฅผ ์ด์šฉํ•˜๋ฉด 25%์ •๋„ ์ ˆ๊ฐํ•  ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค.
12:47
Lastly, I want you to applaud the people who actually create the future,
218
767200
5736
๋งˆ์ง€๋ง‰์œผ๋กœ ์‹ค์ œ๋กœ ๋ฏธ๋ž˜๋ฅผ ๋งŒ๋“ค๊ณ  ์žˆ๋Š” ๋ถ„๋“ค์—๊ฒŒ ๋ฐ•์ˆ˜๋ฅผ ๋ณด๋‚ด์…จ์œผ๋ฉด ์ข‹๊ฒ ์Šต๋‹ˆ๋‹ค.
12:52
Yash Mulgaonkar, Sikang Liu and Giuseppe Loianno,
219
772960
4920
์•ผ์‰ฌ ๋ฌผ๊ฐ€์˜จ์นด, ์‹œ์บ‰ ๋ฆฌ์œ ์™€ ์ฅฌ์„ธํŽ˜ ๋กœ์ด์•„๋…ธ๊ฐ€
12:57
who are responsible for the three demonstrations that you saw.
220
777920
3496
๋ณด์…จ๋˜ ์„ธ ๊ฐ€์ง€ ์‹œ๋ฒ”์„ ๋‹ด๋‹นํ•ด ์ฃผ์…จ์Šต๋‹ˆ๋‹ค.
13:01
Thank you.
221
781440
1176
๊ฐ์‚ฌํ•ฉ๋‹ˆ๋‹ค.
13:02
(Applause)
222
782640
5920
(๋ฐ•์ˆ˜)
์ด ์›น์‚ฌ์ดํŠธ ์ •๋ณด

์ด ์‚ฌ์ดํŠธ๋Š” ์˜์–ด ํ•™์Šต์— ์œ ์šฉํ•œ YouTube ๋™์˜์ƒ์„ ์†Œ๊ฐœํ•ฉ๋‹ˆ๋‹ค. ์ „ ์„ธ๊ณ„ ์ตœ๊ณ ์˜ ์„ ์ƒ๋‹˜๋“ค์ด ๊ฐ€๋ฅด์น˜๋Š” ์˜์–ด ์ˆ˜์—…์„ ๋ณด๊ฒŒ ๋  ๊ฒƒ์ž…๋‹ˆ๋‹ค. ๊ฐ ๋™์˜์ƒ ํŽ˜์ด์ง€์— ํ‘œ์‹œ๋˜๋Š” ์˜์–ด ์ž๋ง‰์„ ๋”๋ธ” ํด๋ฆญํ•˜๋ฉด ๊ทธ๊ณณ์—์„œ ๋™์˜์ƒ์ด ์žฌ์ƒ๋ฉ๋‹ˆ๋‹ค. ๋น„๋””์˜ค ์žฌ์ƒ์— ๋งž์ถฐ ์ž๋ง‰์ด ์Šคํฌ๋กค๋ฉ๋‹ˆ๋‹ค. ์˜๊ฒฌ์ด๋‚˜ ์š”์ฒญ์ด ์žˆ๋Š” ๊ฒฝ์šฐ ์ด ๋ฌธ์˜ ์–‘์‹์„ ์‚ฌ์šฉํ•˜์—ฌ ๋ฌธ์˜ํ•˜์‹ญ์‹œ์˜ค.

https://forms.gle/WvT1wiN1qDtmnspy7