Robots that fly ... and cooperate | Vijay Kumar

2,184,417 views ใƒป 2012-03-01

TED


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

๋ฒˆ์—ญ: Woo Hwang ๊ฒ€ํ† : Bianca Lee
00:20
Good morning.
0
20566
1159
์•ˆ๋…•ํ•˜์„ธ์š”.
00:22
I'm here today to talk about autonomous flying beach balls.
1
22853
4153
์ €๋Š” ์˜ค๋Š˜ ์ž์œจ์ ์œผ๋กœ ๋‚ ์•„๋‹ค๋‹ˆ๋Š”
๋น„์น˜๋ณผ์— ๋Œ€ํ•ด์„œ ๋งํ•˜๋ ค๊ณ  ํ•ฉ๋‹ˆ๋‹ค.
00:27
(Laughter)
2
27030
1008
์•„๋‹™๋‹ˆ๋‹ค, ์ด์™€๊ฐ™์€ ๋ฏผ์ฒฉํ•œ ๋น„ํ–‰ ๋กœ๋ด‡๋“ค์ž…๋‹ˆ๋‹ค.
00:28
No, agile aerial robots like this one.
3
28062
2865
00:31
I'd like to tell you a little bit about the challenges in building these,
4
31624
3460
์ด๋Ÿฐ ๊ธฐ์ˆ ์„ ์ ์šฉ ํ•  ์ˆ˜ ์žˆ๋Š” ๋ช‡๊ฐ€์ง€ ๊ธฐ๋ง‰ํžŒ ๋ฐฉ๋ฒ•๋“ค๊ณผ
์ด๋Ÿฐ ๋กœ๋ด‡๋“ค์„ ์ œ์ž‘ํ•  ๋•Œ ์–ด๋ ค์šด์ ์— ๋Œ€ํ•ด์„œ
00:35
and some of the terrific opportunities for applying this technology.
5
35108
3392
๋งํ•˜๋ ค๊ณ  ํ•ฉ๋‹ˆ๋‹ค.
00:38
So these robots are related to unmanned aerial vehicles.
6
38957
4608
์ด ๋กœ๋ด‡๋“ค์€
๋ฌด์ธ ํ•ญ๊ณต๊ธฐ์™€ ๊ด€๋ จ์ด ์žˆ์Šต๋‹ˆ๋‹ค.
ํ•˜์ง€๋งŒ ์—ฌ๋Ÿฌ๋ถ„์ด ๋ณด๋Š” ์ด ํ•ญ๊ณต๊ธฐ๋“ค์€ ๋„ˆ๋ฌด ํฌ์ฃ .
00:44
However, the vehicles you see here are big.
7
44199
2651
์ˆ˜์ฒœ ํŒŒ์šด๋“œ์— ๋‹ฌํ•˜๋Š” ๋ฌด๊ฒŒ์ด๊ณ ,
00:47
They weigh thousands of pounds, are not by any means agile.
8
47318
3269
์ „ํ˜€ ๋ฏผ์ฒฉํ•˜์ง€ ์•Š์Šต๋‹ˆ๋‹ค.
00:50
They're not even autonomous.
9
50611
1625
์ž์œจ์ ์œผ๋กœ ์›€์ง์ด๋Š” ๊ฒƒ๋„ ์•„๋‹™๋‹ˆ๋‹ค.
00:52
In fact, many of these vehicles are operated by flight crews
10
52886
4182
์‚ฌ์‹ค ๋งŽ์€ ์ด๋Ÿฐ ํ•ญ๊ณต๊ธฐ๋“ค์€
๋‹ค์ˆ˜์˜ ์กฐ์ข…์‚ฌ๊ฐ€ ํฌํ•จ๋œ
์šดํ•ญ ์Šน๋ฌด์›๋“ค๊ณผ
00:57
that can include multiple pilots,
11
57092
2560
00:59
operators of sensors,
12
59676
2206
์ž‘๋™ ์„ผ์„œ๋“ค, ๊ทธ๋ฆฌ๊ณ  ๊ด€์ œ ์‹œ์Šคํ…œ๋“ค์— ์˜ํ•ด
01:01
and mission coordinators.
13
61906
1330
์šดํ•ญ๋˜์–ด ์ง‘๋‹ˆ๋‹ค.
01:03
What we're interested in is developing robots like this --
14
63922
2743
์ €ํฌ๋Š” ์ด๋Ÿฐ ๋กœ๋ด‡๋“ค์„ ๋ฐœ์ „์‹œํ‚ค๋Š” ๋ฐ ๊ด€์‹ฌ์ด ์žˆ๋Š”๋ฐ์š”
-- ์—ฌ๊ธฐ ๋‘๊ฐœ ์‚ฌ์ง„์ด ์žˆ์ง€์š” --
01:06
and here are two other pictures --
15
66689
1690
์—ฌ๋Ÿฌ๋ถ„์ด ํ”ํžˆ ๊ตฌ์ž… ํ•  ์ˆ˜ ์žˆ๋Š” ๋กœ๋ด‡๋“ค์ž…๋‹ˆ๋‹ค.
01:08
of robots that you can buy off the shelf.
16
68403
2683
์ด ๋กœ๋ด‡๋“ค์€ 4๊ฐœ์˜ ๋‚ ๊ฐœ๊ฐ€ ๋‹ฌ๋ฆฐ ํ—ฌ๋ฆฌ์ฝฅํ„ฐ์ด๋ฉฐ,
01:11
So these are helicopters with four rotors,
17
71110
3284
๋Œ€๋žต 1๋ฏธํ„ฐ ์ •๋„ ํฌ๊ธฐ์—
01:14
and they're roughly a meter or so in scale,
18
74418
3690
๋ช‡ ํŒŒ์šด๋“œ์ •๋„ ๋ฌด๊ฒŒ๊ฐ€ ๋‚˜๊ฐ‘๋‹ˆ๋‹ค.
01:18
and weigh several pounds.
19
78132
1460
๊ทธ๋ž˜์„œ ์„ผ์„œ๋“ค๊ณผ ํ”„๋กœ์„ธ์„œ ๋“ฑ์„ ์žฌ์žฅ์ฐฉํ•˜์—ฌ
01:20
And so we retrofit these with sensors and processors,
20
80076
3355
GPS ์—†์ด๋„
01:23
and these robots can fly indoors.
21
83455
2206
์‹ค๋‚ด์—์„œ ๋‚  ์ˆ˜ ์žˆ๋„๋ก ํ–ˆ์Šต๋‹ˆ๋‹ค.
01:25
Without GPS.
22
85685
1319
์ œ์†์— ๋“ค๊ณ  ์žˆ๋Š” ๋กœ๋ด‡์ด
01:27
The robot I'm holding in my hand
23
87457
1888
๋ฐ”๋กœ ๊ทธ ๋กœ๋ด‡์ž…๋‹ˆ๋‹ค.
01:29
is this one,
24
89369
1650
์ด ๋กœ๋ด‡์€ ๋‘ ํ•™์ƒ์ธ,
01:31
and it's been created by two students,
25
91043
3143
์•Œ๋ ‰์Šค์™€ ๋‹ค๋‹ˆ์—˜์ด ๋งŒ๋“ค์—ˆ์ฃ .
01:34
Alex and Daniel.
26
94210
1648
์ด ๋กœ๋ด‡์€ ์•ฝ 0.1ํŒŒ์šด๋“œ(์•ฝ 45g)
01:36
So this weighs a little more than a tenth of a pound.
27
96493
3115
๋ณด๋‹ค ์กฐ๊ธˆ ๋” ๋‚˜๊ฐ‘๋‹ˆ๋‹ค.
01:39
It consumes about 15 watts of power.
28
99632
2579
์•ฝ 15์™€ํŠธ์˜ ์ „๋ ฅ์„ ์†Œ๋ชจํ•˜๊ตฌ์š”.
๊ทธ๋ฆฌ๊ณ  ๋ณด์‹œ๋‹ค์‹œํ”ผ,
01:42
And as you can see, it's about eight inches in diameter.
29
102235
2833
์ง€๋ฆ„์ด ์•ฝ 8์ธ์น˜(์•ฝ 20cm) ์ •๋„ ๋ฉ๋‹ˆ๋‹ค.
์ด ๋กœ๋ด‡๋“ค์ด ์–ด๋–ป๊ฒŒ ์ž‘๋™๋˜๋Š”์ง€
01:46
So let me give you just a very quick tutorial
30
106330
2595
01:48
on how these robots work.
31
108949
1666
๊ฐ„๋‹จํ•˜๊ฒŒ ์„ค๋ช…์„ ๋“œ๋ฆฌ๊ฒ ์Šต๋‹ˆ๋‹ค.
4๊ฐœ์˜ ํšŒ์ „๋‚ ๊ฐœ๊ฐ€ ์žˆ์Šต๋‹ˆ๋‹ค.
01:51
So it has four rotors.
32
111043
1193
01:52
If you spin these rotors at the same speed,
33
112260
2048
๊ฐ™์€ ์†๋„๋กœ ์ด ํšŒ์ „๋‚ ๊ฐœ๋“ค ๋Œ๋ฆฌ๋ฉด,
01:54
the robot hovers.
34
114332
1199
๋กœ๋ด‡์ด ๋ถ€์–‘ํ•˜๊ฒŒ ๋˜์ฃ .
01:56
If you increase the speed of each of these rotors,
35
116420
3636
์ด ๊ฐ๊ฐ์˜ ํšŒ์ „๋‚ ๊ฐœ์˜ ์†๋„๋ฅผ ์˜ฌ๋ฆฌ๋ฉด,
๋กœ๋ด‡์ด ๋‚ ์•„ ์˜ค๋ฅด๊ฒŒ ๋˜๊ณ , ์†๋„๋ฅผ ์˜ฌ๋ฆฌ๊ฒŒ ๋ฉ๋‹ˆ๋‹ค.
02:00
then the robot flies up, it accelerates up.
36
120080
2539
02:02
Of course, if the robot were tilted,
37
122643
2432
๋ฌผ๋ก , ์ˆ˜ํ‰๋ฐฉํ–ฅ์œผ๋กœ
๊ธฐ์šธ์–ด์ง€๋ฉด,
02:05
inclined to the horizontal,
38
125099
1310
02:06
then it would accelerate in this direction.
39
126433
2803
์ด์ชฝ ๋ฐฉํ–ฅ์œผ๋กœ ์†๋„๋ฅผ ์˜ฌ๋ฆฌ๊ฒŒ ๋ฉ๋‹ˆ๋‹ค.
02:09
So to get it to tilt,
40
129686
1413
๋ชธ์ฒด๋ฅผ ๊ธฐ์šธ์ด๊ธฐ ์œ„ํ•ด์„œ๋Š” ํ•œ๊ฐ€์ง€ ๋ฐฉ๋ฒ•์ด ์žˆ๋Š”๋ฐ์š”.
02:11
there's one of two ways of doing it.
41
131123
2093
์ด ๊ทธ๋ฆผ์„ ๋ณด์‹œ๋ฉด,
02:13
So in this picture, you see that rotor four is spinning faster
42
133240
3636
๋„ค๋ฒˆ์งธ ํšŒ์ „๋‚ ๊ฐœ๊ฐ€ ๋” ๋นจ๋ฆฌ ํšŒ์ „ํ•˜๋Š”๊ฑธ ๋ณด์‹ค ์ˆ˜ ์žˆ์ฃ .
02:16
and rotor two is spinning slower.
43
136900
1968
๊ทธ๋ฆฌ๊ณ  ๋‘๋ฒˆ์งธ ํšŒ์ „๋‚ ๊ฐœ๋Š” ๋” ์ฒœ์ฒœํžˆ ๋Œ๊ณ ์žˆ๊ตฌ์š”.
02:18
And when that happens,
44
138892
1560
์ด๋Ÿฐ ์ƒํƒœ๊ฐ€ ๋˜๋ฉด
02:20
there's a moment that causes this robot to roll.
45
140476
3051
์ด ๋กœ๋ด‡์ด ๋น™๋น™ ๋Œ๊ฒŒ ๋˜๋Š”๊ฑฐ์ฃ .
๋‹ค๋ฅธ ๋ฐฉํ–ฅ์œผ๋กœ๋Š”,
02:24
And the other way around,
46
144495
1255
02:25
if you increase the speed of rotor three and decrease the speed of rotor one,
47
145774
5579
์„ธ๋ฒˆ์งธ ๋‚ ๊ฐœ์˜ ์†๋„๋ฅผ ์˜ฌ๋ฆฌ๊ณ ,
์ฒซ๋ฒˆ์งธ ๋‚ ๊ฐœ์˜ ์†๋„๋ฅผ ์ค„์ด๋ฉด,
์•ž์œผ๋กœ ๊ณ ๊ฐœ๋ฅผ ์ˆ™์ด๊ฒŒ ๋˜์ฃ .
02:31
then the robot pitches forward.
48
151377
1859
02:33
And then finally,
49
153820
1168
๊ทธ๋ฆฌ๊ณ  ๋งˆ์ง€๋ง‰์œผ๋กœ,
02:35
if you spin opposite pairs of rotors
50
155012
2894
๋‚ ๊ฐœ ํ•œ ์Œ์„ ๋ฐ˜๋Œ€์ชฝ๋ณด๋‹ค
02:37
faster than the other pair,
51
157930
1692
๋” ๋นจ๋ฆฌ ๋Œ๋ฆฌ๋ฉด,
02:39
then the robot yaws about the vertical axis.
52
159646
2721
๋กœ๋ด‡์€ ์ˆ˜์ง์ถ•์„ ์ค‘์‹ฌ์œผ๋กœ ํ•œ์ชฝ์œผ๋กœ ๊ธฐ์šธ์–ด์ง‘๋‹ˆ๋‹ค.
๋ณด๋“œ์— ํƒ‘์žฌ๋œ ํ”„๋กœ์„ธ์„œ๊ฐ€
02:42
So an on-board processor
53
162391
1460
02:43
essentially looks at what motions need to be executed
54
163875
3307
์–ด๋–ค ๋™์ž‘์ด ์‹คํ–‰ ๋˜์–ด์•ผ ํ•˜๋Š”์ง€๋ฅผ ๊ด€ํ• ํ•˜๊ณ ,
๋™์ž‘๋“ค์„ ์กฐํ•ฉํ•ด์„œ,
02:47
and combines these motions,
55
167206
1970
์ดˆ๋‹น 600๋ฒˆ ์ •๋„๋กœ ๋ชจํ„ฐ์—
02:49
and figures out what commands to send to the motors --
56
169200
3110
์–ด๋–ค ๋ช…๋ น์„ ๋‚ด๋ฆด ๊ฒƒ์ธ์ง€๋ฅผ ๊ฒฐ์ •ํ•ฉ๋‹ˆ๋‹ค.
02:52
600 times a second.
57
172334
1412
02:53
That's basically how this thing operates.
58
173770
1969
์ด๊ฒƒ์ด ๊ธฐ๋ณธ์ ์œผ๋กœ ์ด ๋กœ๋ด‡์ด ์ž‘๋™๋˜๋Š” ๋ฐฉ์‹์ž…๋‹ˆ๋‹ค.
์ด๋Ÿฐ ๋กœ๋ด‡ ์„ค๊ณ„์˜ ์žฅ์ ๋“ค ์ค‘ ํ•œ๊ฐ€์ง€๋Š”
02:56
So one of the advantages of this design
59
176247
2142
๋กœ๋ด‡์˜ ํฌ๊ธฐ๋ฅผ ์ค„์ด๋ฉด
02:58
is when you scale things down,
60
178413
1899
์ž์—ฐ์Šค๋Ÿฝ๊ฒŒ ๋ฏผ์ฒฉํ•ด์ง€์ฃ .
03:00
the robot naturally becomes agile.
61
180336
2318
์—ฌ๊ธฐ "R"์ด
03:03
So here, R is the characteristic length of the robot.
62
183194
4063
๋กœ๋ด‡์˜ ๊ธธ์ด๋ฅผ ํ‘œ์‹œํ•ฉ๋‹ˆ๋‹ค.
์ง€๋ฆ„์˜ ๋ฐ˜์ •๋„์ด์ฃ .
03:07
It's actually half the diameter.
63
187281
1674
03:09
And there are lots of physical parameters that change as you reduce R.
64
189595
4641
R์„ ์ค„์—ฌ๊ฐ€๋ฉด,
๋งŽ์€ ๋ฌผ๋ฆฌ์  ๋ณ€์ˆ˜๋“ค์ด ๋ณ€ํ•˜๊ฒŒ ๋ฉ๋‹ˆ๋‹ค.
03:14
The one that's most important is the inertia,
65
194968
2631
์•„์ฃผ ์ค‘์š”ํ•œ๊ฒƒ ์ค‘ ํ•œ๊ฐ€์ง€๋Š”
๋™์ž‘์— ๋ฏธ์น˜๋Š” ๊ด€์„ฑ์ด๋‚˜ ์ €ํ•ญ๋ ฅ์ž…๋‹ˆ๋‹ค.
03:17
or the resistance to motion.
66
197623
1921
๊ฐ์šด๋™์„ ์ขŒ์šฐํ•˜๋Š” ๊ด€์„ฑ์€
03:19
So it turns out the inertia, which governs angular motion,
67
199568
4476
R์˜ 1/5์”ฉ ๋Š˜์–ด ๋‚œ๋‹ค๊ณ 
์•Œ๋ ค์ ธ์žˆ์Šต๋‹ˆ๋‹ค.
03:24
scales as a fifth power of R.
68
204068
2600
R์˜ ํฌ๊ธฐ๋ฅผ ๋” ์ž‘๊ฒŒํ•˜๋ฉด ํ•  ์ˆ˜๋ก
03:27
So the smaller you make R,
69
207046
1696
03:28
the more dramatically the inertia reduces.
70
208766
2317
๊ด€์„ฑ์€ ๋”์šฑ๋” ๊ธ‰์†ํžˆ ๊ฐ์†Œํ•ฉ๋‹ˆ๋‹ค.
03:31
So as a result, the angular acceleration,
71
211956
2935
๊ฒฐ๊ณผ์ ์œผ๋กœ, ๊ทธ๋ฆฌ์Šค ๋ฌธ์ž "์•ŒํŒŒ"๋กœ ์“ฐ๋Š”
03:34
denoted by the Greek letter alpha here,
72
214915
2039
๊ฐ ๊ฐ€์†๋„๋Š” 1/R๋กœ
03:36
goes as 1 over R.
73
216978
1619
๋Š˜์–ด๋‚ฉ๋‹ˆ๋‹ค.
03:38
It's inversely proportional to R.
74
218621
1729
์ด๊ฒƒ์€ R์— ๋ฐ˜๋น„๋ก€ํ•˜๋Š”๊ฑฐ์ฃ .
03:40
The smaller you make it, the more quickly you can turn.
75
220374
2770
๋” ์ž‘๊ฒŒ ๋งŒ๋“ค๋ฉด ๋งŒ๋“ค ์ˆ˜๋ก ๋” ๋นจ๋ฆฌ ๋ฐฉํ–ฅ์„ ๋ฐ”๊ฟ€ ์ˆ˜ ์žˆ๋Š” ๊ฒƒ์ž…๋‹ˆ๋‹ค.
์ด ๋น„๋””์˜ค์—์„œ ๋” ๋ช…ํ™•ํ•˜๊ฒŒ ๋ณด์—ฌ๋“œ๋ฆฌ๊ฒ ์Šต๋‹ˆ๋‹ค.
03:44
So this should be clear in these videos.
76
224219
1961
์˜ค๋ฅธ์ชฝ ๋ฐ‘์— ๋กœ๋ด‡์„ ๋ณด๋ฉด,
03:46
On the bottom right, you see a robot performing a 360-degree flip
77
226204
4459
0.5์ดˆ๋„ ์•ˆ๋˜๋Š” ์‹œ๊ฐ„๋งŒ์—
03:50
in less than half a second.
78
230687
1548
360๋„ ๋’ค์ง‘๊ธฐ๋ฅผ ํ•ฉ๋‹ˆ๋‹ค.
03:52
Multiple flips, a little more time.
79
232656
2580
๋ช‡๋ฒˆ ๋” ๋‹ค์ค‘ ๋’ค์ง‘๊ธฐ๋ฅผ ํ•ฉ๋‹ˆ๋‹ค.
์—ฌ๊ธฐ ํ”„๋กœ์„ธ์„œ๋“ค์ด
03:56
So here the processes on board
80
236259
1881
๊ฐ€์†๊ณ„๋กœ๋ถ€ํ„ฐ ์ •๋ณด๋ฅผ ๋ฐ›๊ณ 
03:58
are getting feedback from accelerometers and gyros on board,
81
238164
3184
์ด์ „์— ๋งํ•œ๋Œ€๋กœ
04:01
and calculating, like I said before,
82
241372
2118
์ดˆ๋‹น 600๋ฒˆ ์ •๋„
04:03
commands at 600 times a second,
83
243514
1809
ํ‰ํ–‰ ํšŒ์ „์ž๊ฐ€ ๊ณ„์‚ฐํ•ด์„œ
04:05
to stabilize this robot.
84
245347
1889
๋กœ๋ด‡์ด ํ‰ํ–‰์„ ์œ ์ง€ํ•˜๋„๋ก ํ•ฉ๋‹ˆ๋‹ค.
04:07
So on the left, you see Daniel throwing this robot up into the air,
85
247607
3350
์™ผ์ชฝ์—, ๋‹ค๋‹ˆ์—˜์ด ๋กœ๋ด‡์„ ํ•˜๋Š˜๋กœ ๋‚ ๋ฆฌ๋Š” ๋ชจ์Šต์„ ๋ณผ ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค.
04:10
and it shows you how robust the control is.
86
250981
2027
์ œ์–ด๊ฐ€ ์–ผ๋งˆ๋‚˜ ์ž˜ ๋˜๋Š”์ง€ ๋ณด์—ฌ์ฃผ์ฃ .
์–ด๋–ป๊ฒŒ ๋˜์ง€๋”๋ผ๋„,
04:13
No matter how you throw it,
87
253032
1310
04:14
the robot recovers and comes back to him.
88
254366
2912
๋กœ๋ด‡์€ ๋ณต๊ตฌํ•˜์—ฌ ๋‹ค๋‹ˆ์—˜์—๊ฒŒ ๋Œ์•„๊ฐ‘๋‹ˆ๋‹ค.
04:18
So why build robots like this?
89
258881
1777
๊ทธ๋Ÿผ ์™œ ์ด๋Ÿฐ ๋กœ๋ด‡์„ ๋งŒ๋“ค์—ˆ์„๊นŒ์š”?
์ด๋Ÿฐ ๋กœ๋ด‡์€ ๋งŽ์€ ์‘์šฉ๋ถ„์•ผ์— ์‚ฌ์šฉ ๋  ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค.
04:21
Well, robots like this have many applications.
90
261079
2571
์ด์ฒ˜๋Ÿผ ์นจ์ž…์ž๋ฅผ ํƒ์ง€ํ•˜๊ธฐ ์œ„ํ•ด
04:24
You can send them inside buildings like this,
91
264198
2471
04:26
as first responders to look for intruders,
92
266693
3587
๊ฑด๋ฌผ ๋‚ด๋ถ€๋กœ ๋ณด๋‚ผ ์ˆ˜ ๋„ ์žˆ๊ณ ,
์ƒํ™”ํ•™ ๋ฌผ์งˆ ์œ ์ถœ์ด๋‚˜
04:30
maybe look for biochemical leaks,
93
270304
3301
๊ฐ€์Šค ์œ ์ถœ์„ ํƒ์ง€ํ•˜๊ธฐ ์œ„ํ•ด์„œ ์‚ฌ์šฉ ๋  ์ˆ˜ ๋„ ์žˆ์ฃ .
04:33
gaseous leaks.
94
273629
1151
๊ฑด์„ค๊ณผ ๊ฐ™์€ ์‘์šฉ๋ถ„์•ผ์—์„œ๋„
04:35
You can also use them for applications like construction.
95
275102
3714
์‚ฌ์šฉ ๋  ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค.
04:38
So here are robots carrying beams, columns
96
278840
4524
๋น”์„ ์šด๋ฐ˜ํ•˜๊ฑฐ๋‚˜, ์œก๋ฉด์ฒด์˜ ๊ตฌ์กฐ๋ฌผ์„
๋‚˜์—ดํ•˜๊ฑฐ๋‚˜ ์กฐํ•ฉ ํ•  ์ˆ˜ ๋„ ์žˆ์Šต๋‹ˆ๋‹ค.
04:43
and assembling cube-like structures.
97
283388
1848
04:45
I'll tell you a little bit more about this.
98
285260
2056
์ด๋ถ€๋ถ„์— ๋Œ€ํ•ด ์ข€๋” ์ด์•ผ๊ธฐ ํ•ด ๋ณด๊ฒ ์Šต๋‹ˆ๋‹ค.
04:48
The robots can be used for transporting cargo.
99
288695
2541
๋กœ๋ด‡์€ ํ™”๋ฌผ ์šด์†ก์„ ์œ„ํ•ด์„œ ์‚ฌ์šฉ ๋  ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค.
04:51
So one of the problems with these small robots
100
291583
3113
์ด๋Ÿฐ ์ž‘์€ ๋กœ๋ด‡๋“ค์˜ ๋ฌธ์ œ์ค‘ ํ•œ๊ฐ€์ง€๋Š”
04:54
is their payload-carrying capacity.
101
294720
2001
์šด๋ฐ˜๊ฐ€๋Šฅํ•œ ์šฉ๋Ÿ‰์˜ ์ ์žฌํ•˜์ค‘์ž…๋‹ˆ๋‹ค.
04:56
So you might want to have multiple robots carry payloads.
102
296745
3017
๊ทธ๋ž˜์„œ ํƒ‘์žฌ๋Ÿ‰์„ ์œ„ํ•ด
์—ฌ๋Ÿฌ๋Œ€์˜ ๋กœ๋ด‡๋“ค์„ ์‚ฌ์šฉ ํ•  ์ˆ˜ ๋„ ์žˆ๊ฒ ์ฃ .
05:00
This is a picture of a recent experiment we did --
103
300746
2355
์ด ๊ทธ๋ฆผ์ด ์ตœ๊ทผ์— ์‹คํ—˜ํ–ˆ๋˜ ๋ชจ์Šต๋‹ˆ๋‹ค.
-- ์‚ฌ์‹ค ๋”์ด์ƒ ์ตœ๊ทผ์€ ์•„๋‹ˆ์ง€๋งŒ์š” --
05:03
actually not so recent anymore --
104
303125
1595
05:04
in Sendai, shortly after the earthquake.
105
304744
2492
์ง€์ง„ํ›„์˜ ์„ผ๋‹ค์ด ์ง€๋ฐฉ์ž…๋‹ˆ๋‹ค.
05:07
So robots like this could be sent into collapsed buildings,
106
307938
3191
๋กœ๋ด‡๋“ค์€ ์ฒœ์žฌ์ง€๋ณ€ ์žฌ๋‚œ ์ดํ›„์— ์œ„ํ—˜๋ฌผ์— ์ ‘๊ทผํ•˜๋ ค๊ณ 
๋ถ•๊ดด๋œ ๊ฑด๋ฌผ๋“ค ์•ˆ์œผ๋กœ ๋“ค์–ด ๊ฐˆ ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค.
05:11
to assess the damage after natural disasters,
107
311153
2891
๋˜๋Š” ๋ฐฉ์‚ฌ๋Šฅ ์ˆ˜์น˜๋ฅผ ๊ทธ๋ฆฌ๊ธฐ ์œ„ํ•ด
05:14
or sent into reactor buildings,
108
314068
1582
05:15
to map radiation levels.
109
315674
1619
์›์ž๋กœ ๊ฑด๋ฌผ๋กœ ๋“ค์–ด ๊ฐˆ ์ˆ˜ ๋„ ์žˆ๊ตฌ์š”.
05:19
So one fundamental problem that the robots have to solve
110
319583
3521
๊ทธ๋ž˜์„œ ์ด๋Ÿฐ ๋กœ๋ด‡๋“ค์ด ํ’€์–ด์•ผ๋งŒ ํ•˜๋Š”
์•„์ฃผ ๊ธฐ๋ณธ์ ์ธ ๋ฌธ์ œ์ ์€
05:23
if they are to be autonomous,
111
323128
1579
05:24
is essentially figuring out how to get from point A to point B.
112
324731
3505
์ž์œจ์ ์ธ ๋กœ๋ด‡์ด๋ผ๋ฉด
A์ง€์ ์—์„œ B์ง€์ ์œผ๋กœ ์–ด๋–ป๊ฒŒ ๊ฐ€๋Š๋ƒ ์ž…๋‹ˆ๋‹ค.
05:28
So this gets a little challenging,
113
328937
1667
์ด๊ฑด ์กฐ๊ธˆ ์–ด๋ ค์šด ๋ฌธ์ œ์ธ๋ฐ์š”,
05:30
because the dynamics of this robot are quite complicated.
114
330628
3007
์ด ๋กœ๋ด‡์˜ ์—ญํ•™๊ตฌ์กฐ๊ฐ€ ๋ณต์žกํ•˜๊ธฐ ๋•Œ๋ฌธ์ž…๋‹ˆ๋‹ค.
05:33
In fact, they live in a 12-dimensional space.
115
333659
2492
์ด ๋กœ๋ด‡๋“ค์€ 12์ฐจ์›์˜ ๊ณต๊ฐ„์—์„œ ์‚ฐ๋‹ค๊ณ  ๋ณผ ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค.
๊ทธ๋ž˜์„œ ์•ฝ๊ฐ„์˜ ํŠธ๋ฆญ์„ ์‚ฌ์šฉํ•ฉ๋‹ˆ๋‹ค.
05:36
So we use a little trick.
116
336524
1444
05:37
We take this curved 12-dimensional space,
117
337992
3413
์ด ๊ตฌ๋ถˆ์–ด์ง„ 12์ฐจ์› ๊ณต๊ฐ„์„
ํ‰ํ‰ํ•œ 4์ฐจ์› ๊ณต๊ฐ„์œผ๋กœ
05:41
and transform it into a flat, four-dimensional space.
118
341429
3894
๋ณ€ํ˜•ํ•ฉ๋‹ˆ๋‹ค.
4์ฐจ์› ๊ณต๊ฐ„์€
05:45
And that four-dimensional space consists of X, Y, Z,
119
345347
3071
X, Y, Z, ๊ทธ๋ฆฌ๊ณ  ์š”์ž‰์ถ•์œผ๋กœ ๊ตฌ์„ฑ๋ฉ๋‹ˆ๋‹ค.
05:48
and then the yaw angle.
120
348442
1412
05:49
And so what the robot does,
121
349878
1358
๊ทธ๋ฆฌ๊ณ  ๋กœ๋ด‡์ด ํ•˜๋Š”๊ฒƒ์€
05:51
is it plans what we call a minimum-snap trajectory.
122
351260
3976
์ตœ์†Œ ๋ชฉํ‘œ ๊ถค์ ์ด๋ผ๊ณ  ๋ถˆ๋ฆฌ๋Š” ๊ถค์ ์„ ๊ณ„์‚ฐํ•ฉ๋‹ˆ๋‹ค.
๋ฌผ๋ฆฌํ•™์œผ๋กœ ์•Œ๋ ค๋“œ๋ฆฌ๋ฉด,
05:56
So to remind you of physics:
123
356209
1397
05:57
You have position, derivative, velocity;
124
357630
2054
์œ„์น˜, ๋ฏธ๋ถ„, ์†๋ ฅ,
05:59
then acceleration;
125
359708
1856
๊ทธ๋ฆฌ๊ณ  ๊ฐ€์†๋„ ์ž…๋‹ˆ๋‹ค.
06:01
and then comes jerk,
126
361588
2101
๊ทธ๋ž˜์„œ ๊ฐ‘์ž๊ธฐ ์›€์ง์ด๊ณ 
06:03
and then comes snap.
127
363713
1420
์ˆœ๊ฐ„์ ์œผ๋กœ ํ•˜๊ฐ•ํ•ฉ๋‹ˆ๋‹ค..
06:05
So this robot minimizes snap.
128
365578
2452
๊ทธ๋ž˜์„œ ์ด ๋กœ๋ด‡์€ ์ˆœ๊ฐ„ํ•˜๊ฐ•์„ ์ตœ์†Œํ™”ํ•ฉ๋‹ˆ๋‹ค.
06:08
So what that effectively does,
129
368800
1436
๊ทธ๋ž˜์„œ ๋กœ๋ด‡์ด ํšจ๊ณผ์ ์œผ๋กœ ํ•˜๋Š” ๊ฒƒ์€
06:10
is produce a smooth and graceful motion.
130
370260
2624
๋ถ€๋“œ๋Ÿฝ๊ณ  ์šฐ์•„ํ•œ ๋™์ž‘์„ ํ•˜๋Š”๊ฒƒ ์ž…๋‹ˆ๋‹ค.
06:12
And it does that avoiding obstacles.
131
372908
2842
๊ทธ๋ž˜์„œ ์žฅ์• ๋ฌผ์„ ํ”ผํ•˜๊ธฐ ์œ„ํ•œ ๊ฒƒ์ด์ฃ .
์ด๋Ÿฐ ํ‰ํ‰ํ•œ ๊ณต๊ฐ„์—์„œ ์ตœ์†Œ ๋ชฉํ‘œ ๊ถค์ ๋“ค์€
06:16
So these minimum-snap trajectories in this flat space are then transformed
132
376313
3563
๋กœ๋ด‡๋“ค์ด ์ œ์–ดํ•˜๊ณ 
06:19
back into this complicated 12-dimensional space,
133
379900
3325
์‹คํ–‰ํ•ด์•ผํ•˜๋Š”
๋ณต์žกํ•œ 12์ฐจ์› ๊ณต๊ฐ„์œผ๋กœ
06:23
which the robot must do for control and then execution.
134
383249
2986
๋‹ค์‹œ ์žฌ๋ณ€ํ˜•ํ•˜๊ฒŒ ๋ฉ๋‹ˆ๋‹ค.
06:26
So let me show you some examples
135
386749
1540
์ด๋Ÿฐ ์ตœ์†Œ ๋ชฉํ‘œ ๊ถค์ ๋“ค์ด
06:28
of what these minimum-snap trajectories look like.
136
388313
2777
์–ด๋–ค๊ฒƒ์ธ์ง€ ๋ช‡๊ฐ€์ง€ ์˜ˆ๋ฅผ ๋ณด์—ฌ ๋“œ๋ฆฌ ๊ฒ ์Šต๋‹ˆ๋‹ค.
์ฒซ๋ฒˆ์งธ ๋น„๋””์˜ค์—์„œ,
06:31
And in the first video,
137
391114
1151
06:32
you'll see the robot going from point A to point B,
138
392289
2729
์ค‘๊ฐ„์ง€์ ์„ ํ†ต๊ณผํ•ด์„œ
A์ง€์ ์—์„œ B์ง€์ ์œผ๋กœ ์ด๋™ํ•˜๋Š” ๋กœ๋ด‡์„ ๋ณด์‹œ๊ฒŒ๋ฉ๋‹ˆ๋‹ค.
06:35
through an intermediate point.
139
395042
1623
06:36
(Whirring noise)
140
396990
2730
๊ทธ๋ž˜์„œ ์ด ๋กœ๋ด‡์ด ๋ถ„๋ช…ํ•˜๊ฒŒ ๊ณก์„  ๊ถค์ ์„
06:43
So the robot is obviously capable of executing any curve trajectory.
141
403377
3621
๋”ฐ๋ผ ๊ฐˆ ์ˆ˜ ์žˆ๋‹ค๋Š” ๊ฒƒ์„ ๋ณผ ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค.
๋กœ๋ด‡์€ ์ค‘๋ ฅ์˜ 2๋ฐฐ๋ฅผ ์ด๊ฒจ๋‚ด๊ณ 
06:47
So these are circular trajectories,
142
407022
1714
06:48
where the robot pulls about two G's.
143
408760
2500
ํƒ€์›๊ถค์ ์„ ๋Œ๊ธฐ๋„ ํ•ฉ๋‹ˆ๋‹ค.
06:52
Here you have overhead motion capture cameras on the top
144
412844
3702
์ดˆ๋‹น 100๋ฒˆ์ •๋„ ๋กœ๋ด‡์ด ์–ด๋”” ์žˆ๋Š”์ง€๋ฅผ ํ™•์ธ์‹œ์ผœ์ฃผ๋Š”
06:56
that tell the robot where it is 100 times a second.
145
416570
3206
๋ชจ์…˜์บก์ณ ์นด๋ฉ”๋ผ๋ฅผ ์ฒœ์žฅ์— ๋‹ฌ์•˜์Šต๋‹ˆ๋‹ค.
06:59
It also tells the robot where these obstacles are.
146
419800
2436
์ด ์นด๋ฉ”๋ผ๋Š” ๋กœ๋ด‡์—๊ฒŒ ์žฅ์• ๋ฌผ์ด ์–ด๋””์— ์žˆ๋Š”์ง€๋„ ์•Œ๋ ค์ค๋‹ˆ๋‹ค.
์žฅ์• ๋ฌผ์€ ์›€์ง์ผ์ˆ˜๋„ ์žˆ๊ตฌ์š”.
07:03
And the obstacles can be moving.
147
423140
1556
07:04
And here, you'll see Daniel throw this hoop into the air,
148
424720
3127
๋‹ค๋‹ˆ์—˜์ด ํ—ˆ๊ณต์— ํ›Œ๋ผํ›„ํ”„๋ฅผ ๋˜์ง€๋Š” ๊ฒƒ์„ ๋ณผ ์ˆ˜ ์žˆ๋Š”๋ฐ์š”,
07:07
while the robot is calculating the position of the hoop,
149
427871
2659
๋กœ๋ด‡์ด ํ›Œ๋ผํ›„ํ”„์˜ ์œ„์น˜๋ฅผ ๊ณ„์‚ฐํ•˜๊ณ ,
ํ›Œ๋ผํ›„ํ”„๋ฅผ ์ž˜ ํ†ต๊ณผ ํ•˜๋„๋ก ์‹œ๋„ํ•˜๋Š” ๊ฒƒ์„ ๋ณผ ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค.
07:10
and trying to figure out how to best go through the hoop.
150
430554
2880
ํ•™์ž๋กœ์จ, ์—ฐ๊ตฌ์ž๊ธˆ์„ ํ™•๋ณด ํ•˜๊ธฐ ์œ„ํ•ด์„œ
07:14
So as an academic,
151
434002
1234
07:15
we're always trained to be able to jump through hoops
152
435260
2528
ํ›Œ๋ผํ›„ํ”„๋ฅผ ๋›ฐ์–ด๋„˜๋Š” ํ›ˆ๋ จ์„ ํ•ด์™”์Šต๋‹ˆ๋‹ค.
07:17
to raise funding for our labs,
153
437812
1485
๊ทธ๋ฆฌ๊ณ  ์šฐ๋ฆฌ์˜ ๋กœ๋ด‡๋“ค์ด ํ•ด๋ƒˆ์Šต๋‹ˆ๋‹ค.
07:19
and we get our robots to do that.
154
439321
1915
07:21
(Applause)
155
441260
6365
(๋ฐ•์ˆ˜)
๋กœ๋ด‡์ด ํ•  ์ˆ˜ ์žˆ๋Š” ๋˜๋‹ค๋ฅธ ๊ฒƒ์€
07:28
So another thing the robot can do
156
448524
1611
๋ฏธ๋ฆฌ ํ”„๋กœ๊ทธ๋žจ ๋˜๊ฑฐ๋‚˜ ํ•™์Šต๋œ ๊ถค๋„๋ฅผ
07:30
is it remembers pieces of trajectory
157
450159
2077
07:32
that it learns or is pre-programmed.
158
452260
2904
๊ธฐ์–ต ํ•  ์ˆ˜ ์žˆ๋‹ค๋Š” ๊ฒƒ์ž…๋‹ˆ๋‹ค.
์šด๋™๋Ÿ‰์„ ๋†’์—ฌ์„œ
07:35
So here, you see the robot combining a motion that builds up momentum,
159
455466
4770
๋™์ž‘์„ ์กฐํ•ฉํ•˜๊ณ ,
๋ฐฉํ–ฅ์„ ๋ฐ”๊พธ๊ณ 
07:40
and then changes its orientation and then recovers.
160
460260
2976
๋‹ค์‹œ ๋ณต๊ท€ํ•˜๋Š” ๋กœ๋ด‡์„ ๋ณผ ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค.
์ฐฝ๋ฌธ์‚ฌ์ด์˜ ํ‹ˆ์ด ๋กœ๋ด‡๋ณด๋‹ค ์•ฝ๊ฐ„ ๋” ํฌ๊ธฐ ๋•Œ๋ฌธ์—
07:44
So it has to do this because this gap in the window
161
464004
2960
07:46
is only slightly larger than the width of the robot.
162
466988
3458
์ด ๋กœ๋ด‡์€ ์ด๋Ÿฐ ๋™์ž‘์„ ํ•ด์•ผ๋งŒ ํ•ฉ๋‹ˆ๋‹ค.
๊ทธ๋ž˜์„œ ๋„์•ฝํŒ ์œ„์— ์„œ ์žˆ๋Š” ๋‹ค์ด๋ฒ„ ์ฒ˜๋Ÿผ
07:51
So just like a diver stands on a springboard
163
471050
2753
07:53
and then jumps off it to gain momentum,
164
473827
2453
์šด๋™๋Ÿ‰์„ ์–ป๊ธฐ ์œ„ํ•ด์„œ ๋›ฐ์–ด ๋‚ด๋ฆฝ๋‹ˆ๋‹ค.
๊ทธ๋ž˜์„œ ๋ฐœ๋ ˆ์ฒ˜๋Ÿผ ํ”ผ๋ฃจ์—ฃ ๋Œ๊ธฐ๋‚˜ ๊ณต์ค‘์ œ๋น„๋ฅผ ๋Œ๊ณ ,
07:56
and then does this pirouette, this two and a half somersault through
165
476304
3229
๊ทธ๋ฆฌ๊ณ  ์šฐ์•„ํ•˜๊ฒŒ ๋‹ค์‹œ ๋ณต๊ท€ํ•˜๊ณ ,
07:59
and then gracefully recovers,
166
479557
1402
08:00
this robot is basically doing that.
167
480983
1698
์ด ๋กœ๋ด‡์ด ๊ธฐ๋ณธ์ ์œผ๋กœ ํ•˜๋Š” ๊ฒƒ์ž…๋‹ˆ๋‹ค.
08:02
So it knows how to combine little bits and pieces of trajectories
168
482705
3080
๊ทธ๋ž˜์„œ ๋กœ๋ด‡์€ ์ƒ๋‹นํžˆ ๋ณต์žกํ•œ ์ž„๋ฌด๋ฅผ ์ˆ˜ํ–‰ํ•˜๊ธฐ ์œ„ํ•ด
08:05
to do these fairly difficult tasks.
169
485809
2999
์–ด๋–ป๊ฒŒ ๊ถค๋„๋ฅผ ์กฐํ•ฉํ•ด์•ผํ•˜๋Š”์ง€ ์•Œ๊ณ  ์žˆ์Šต๋‹ˆ๋‹ค.
์ €๋Š” ์žฅ์น˜๋“ค์„ ๋ฐ”๊พธ๊ณ  ์‹ถ์Šต๋‹ˆ๋‹ค.
08:10
So I want change gears.
170
490018
1218
08:11
So one of the disadvantages of these small robots is its size.
171
491260
3839
์ด ์ž‘์€ ๋กœ๋ด‡๋“ค์˜ ๋‹จ์  ์ค‘ ํ•œ๊ฐ€์ง€๋Š” ํฌ๊ธฐ์ž…๋‹ˆ๋‹ค.
์ฒ˜์Œ์— ๋ง์”€๋“œ๋ฆฐ๋Œ€๋กœ,
08:15
And I told you earlier
172
495654
1152
08:16
that we may want to employ lots and lots of robots
173
496830
2349
ํฌ๊ธฐ ๋ฌธ์ œ๋ฅผ ๊ทน๋ณตํ•˜๊ธฐ ์œ„ํ•ด์„œ
๋” ๋งŽ์€ ๋กœ๋ด‡๋“ค์„ ์ด์šฉํ•˜๊ธฐ๋ฅผ ์›ํ•œ๋‹ค๊ณ  ํ–ˆ์Šต๋‹ˆ๋‹ค.
08:19
to overcome the limitations of size.
174
499203
2033
ํ•œ๊ฐ€์ง€ ์–ด๋ ค์šด์ ์€
08:22
So one difficulty is:
175
502010
1574
08:23
How do you coordinate lots of these robots?
176
503608
2793
์–ด๋–ป๊ฒŒ ์ด๋Ÿฐ ๋กœ๋ด‡๋“ค์„ ์กฐ์ง์ ์œผ๋กœ ์›€์ง์ด๋Š๋ƒ ์ž…๋‹ˆ๋‹ค.
08:26
And so here, we looked to nature.
177
506425
1811
๊ทธ๋ž˜์„œ ์ž์—ฐ์—์„œ ์ฐพ์•„๋ณด์•˜์Šต๋‹ˆ๋‹ค.
08:28
So I want to show you a clip of Aphaenogaster desert ants,
178
508544
4608
์Šคํ‹ฐ๋ธ ํ”„๋ž ๊ต์ˆ˜์˜ ์—ฐ๊ตฌ์‹ค์—์„œ ๊ฐ€์ ธ์˜จ
์‚ฌ๋ง‰๊ฐœ๋ฏธ์˜ ๋™์˜์ƒ์„
๋ณด์—ฌ๋“œ๋ฆฌ๋„๋ก ํ•˜๊ฒ ์Šต๋‹ˆ๋‹ค.
08:33
in Professor Stephen Pratt's lab, carrying an object.
179
513176
3388
์ด๊ฒƒ์€ ๋ฌดํ™”๊ณผ ์กฐ๊ฐ์ž…๋‹ˆ๋‹ค.
08:36
So this is actually a piece of fig.
180
516588
1715
๋ฌดํ™”๊ณผ ์ฆ™์„ ๋ฐ”๋ฅธ ๋‹ค๋ฅธ ๋ฌผ๊ฑด์œผ๋กœ๋„ ํ•  ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค.
08:38
Actually you take any object coated with fig juice,
181
518327
2400
๊ฐœ๋ฏธ๋“ค์ด ์ด ๋ฌผ๊ฑด์„ ๊ฐœ๋ฏธ์ง‘์œผ๋กœ ์˜ฎ๊ธธ๊ฒ๋‹ˆ๋‹ค.
08:40
and the ants will carry it back to the nest.
182
520751
2091
08:42
So these ants don't have any central coordinator.
183
522866
3169
์ด ๊ฐœ๋ฏธ๋“ค์€ ์–ด๋–ค ์ค‘์•™ ๊ด€๋ฆฌ์ž๋„ ์—†์Šต๋‹ˆ๋‹ค.
๊ฐœ๋ฏธ๋“ค์€ ์ž์‹ ๋“ค์˜ ์ด์›ƒ์„ ๊ฐ์ง€ํ•ฉ๋‹ˆ๋‹ค.
08:46
They sense their neighbors.
184
526487
1547
๋ช…ํ™•ํ•œ ์˜์‚ฌ์†Œํ†ต์€ ์—†์Šต๋‹ˆ๋‹ค.
08:48
There's no explicit communication.
185
528058
2285
ํ•˜์ง€๋งŒ ์ด์›ƒ๋“ค์„ ๊ฐ์ง€ํ•˜๊ธฐ ๋•Œ๋ฌธ์—,
08:50
But because they sense the neighbors
186
530367
1739
๊ทธ๋ฆฌ๊ณ  ๋ฌผ๊ฑด์„ ๊ฐ์ง€ํ•˜๊ธฐ ๋•Œ๋ฌธ์—
08:52
and because they sense the object,
187
532130
1776
08:53
they have implicit coordination across the group.
188
533930
2830
๊ทธ๋ฃน ๋‚ด์— ๋ฌต์‹œ์ ์ธ ์กฐ์งํ™”๊ฐ€ ์ด๋ฃจ์–ด์ง‘๋‹ˆ๋‹ค.
์ด๋Ÿฐ๊ฒƒ์ด ์šฐ๋ฆฌ์˜ ๋กœ๋ด‡๋“ค์ด ๊ฐ–๊ธฐ๋ฅผ ๋ฐ”๋ผ๋Š”
08:57
So this is the kind of coordination we want our robots to have.
189
537474
3420
์กฐ์งํ™”์ผ ์ผ์ข…์ž…๋‹ˆ๋‹ค.
09:01
So when we have a robot which is surrounded by neighbors --
190
541569
4778
์ฃผ์œ„ ๋กœ๋ด‡๋“ค๋กœ ๋‘˜๋Ÿฌ์‹ธ์—ฌ์žˆ๋Š”
๋กœ๋ด‡์ด ์žˆ์„ ๋•Œ,
-- ๋กœ๋ด‡ I์™€ ๋กœ๋ด‡ J๋ฅผ ๋ณด์‹œ์ฃ . --
09:06
and let's look at robot I and robot J --
191
546371
2229
์šฐ๋ฆฌ๊ฐ€ ๊ทธ ๋กœ๋ด‡๋“ค์—๊ฒŒ ์›ํ•˜๋Š” ๊ฒƒ์€
09:08
what we want the robots to do,
192
548624
1548
ํฌ๋ฉ”์ด์…˜์„ ํ˜•์„ฑํ•˜๊ณ  ๋น„ํ–‰ํ•˜๋ฉด์„œ
09:10
is to monitor the separation between them,
193
550196
2111
09:12
as they fly in formation.
194
552331
1905
๋กœ๋ด‡๋“ค์ด ๋ถ„๋ฆฌ๋˜์–ด ๋น„ํ–‰ํ•˜๋Š”์ง€ ๊ฐ์‹œํ•˜๋Š” ๊ฒƒ ์ž…๋‹ˆ๋‹ค.
09:14
And then you want to make sure
195
554537
1481
์ด๋Ÿฐ ๋ถ„๋ฆฌ ๋น„ํ–‰์ด ๊ฐ๋‹นํ•  ๋งŒํ•œ ์ˆ˜์ค€์ธ์ง€๋ฅผ
09:16
that this separation is within acceptable levels.
196
556042
2904
ํ™•์ธํ•˜๊ณ  ์‹ถ์–ด์ง€์‹ค๊ฒ๋‹ˆ๋‹ค.
๋กœ๋ด‡๋“ค์€ ์ด๋Ÿฐ ์˜ค๋ฅ˜๋ฅผ ๊ฐ์‹œํ•˜๊ณ 
09:19
So again, the robots monitor this error
197
559303
2468
09:21
and calculate the control commands 100 times a second,
198
561795
4149
์ œ์–ด ๋ช…๋ น์–ด๋ฅผ ์ดˆ๋‹น 100๋ฒˆ์”ฉ
๊ณ„์‚ฐํ•ฉ๋‹ˆ๋‹ค.
09:25
which then translates into motor commands,
199
565968
2023
๊ทธ๋ฆฌ๊ณ  ๋‚˜์„œ ์ดˆ๋‹น 600๋ฒˆ์˜ ๋ชจํ„ฐ ๋ช…๋ น์–ด๋กœ ๋ฐ”๊ฟ‰๋‹ˆ๋‹ค.
09:28
600 times a second.
200
568015
1215
๊ทธ๋ž˜์„œ ์ด๋ฐฉ๋ฒ•์€
09:29
So this also has to be done in a decentralized way.
201
569254
3222
๋น„์ค‘์•™์‹ ๋ฐฉ๋ฒ•์œผ๋กœ ์ด๋ฃจ์–ด์ง‘๋‹ˆ๋‹ค.
09:32
Again, if you have lots and lots of robots,
202
572785
2334
์ˆ˜๋งŽ์€ ๋กœ๋ด‡์„ ๊ฐ€์ง€๊ณ  ์žˆ๋‹ค๋ฉด,
์ด๋Ÿฐ ์ž‘์—…์„ ์ˆ˜ํ–‰ํ•˜๊ธฐ ์œ„ํ•ด ๋ชจ๋“  ์ •๋ณด๋ฅผ ์ค‘์•™์—์„œ
09:35
it's impossible to coordinate all this information centrally
203
575143
3277
09:38
fast enough in order for the robots to accomplish the task.
204
578444
3177
๋นจ๋ฆฌ ์ฒ˜๋ฆฌ ํ•œ๋‹ค๋Š” ๊ฒƒ์€ ๋ถˆ๊ฐ€๋Šฅํ•ฉ๋‹ˆ๋‹ค.
09:41
Plus, the robots have to base their actions only on local information --
205
581645
4348
๋˜ํ•œ ๋กœ๋ด‡๋“ค์€ ๊ทธ๋“ค์˜ ์ด์›ƒ ๋กœ๋ด‡๋“ค๋กœ๋ถ€ํ„ฐ
๊ฐ์ง€ํ•œ ์ง€์—ญ ์ •๋ณด๋งŒ์„
๊ธฐ๋ฐ˜์œผ๋กœ ํ–‰๋™ํ•ด์•ผ๋งŒ ํ•ฉ๋‹ˆ๋‹ค.
09:46
what they sense from their neighbors.
206
586017
2119
๊ทธ๋ฆฌ๊ณ  ๋งˆ์ง€๋ง‰์œผ๋กœ,
09:48
And then finally,
207
588160
1386
09:49
we insist that the robots be agnostic to who their neighbors are.
208
589570
4100
์šฐ๋ฆฌ๋Š” ๋กœ๋ด‡๋“ค์ด
์ด์›ƒ ๋กœ๋ด‡๋“ค๋กœ๋ถ€ํ„ฐ ์ž์œ ๋กญ๋‹ค๊ณ  ๋งํ•ฉ๋‹ˆ๋‹ค.
09:53
So this is what we call anonymity.
209
593694
2117
์šฐ๋ฆฌ๋Š” ์ด๊ฒƒ์„ ์ต๋ช…์„ฑ์ด๋ผ๊ณ  ๋ถ€๋ฆ…๋‹ˆ๋‹ค.
๋‹ค์Œ์— ๋ณด์—ฌ๋“œ๊ณ  ์‹ถ์€ ๊ฒƒ์€
09:57
So what I want to show you next is a video of 20 of these little robots,
210
597258
6417
20๊ฐœ์˜ ์ž‘์€ ๋กœ๋ด‡๋“ค์ด
๋ฌด๋ฆฌ์ง€์–ด ๋‚ ์•„๋‹ค๋‹ˆ๋Š”
10:03
flying in formation.
211
603699
1537
๋น„๋””์˜ค ์ž…๋‹ˆ๋‹ค.
๋กœ๋ด‡๋“ค์€ ๊ทธ๋“ค ์ฃผ๋ณ€ ๋กœ๋ด‡๋“ค์˜ ์œ„์น˜๋ฅผ ๊ฐ์‹œํ•ฉ๋‹ˆ๋‹ค.
10:06
They're monitoring their neighbors' positions.
212
606144
2896
๊ทธ๋ฆฌ๊ณ  ํฌ๋ฉ”์ด์…˜์„ ์œ ์ง€ํ•ฉ๋‹ˆ๋‹ค.
10:09
They're maintaining formation.
213
609064
1902
10:10
The formations can change.
214
610990
1780
ํฌ๋ฉ”์ด์…˜์€ ๋ฐ”๊ฟ€ ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค.
10:12
They can be planar formations,
215
612794
1679
ํ‰๋ฉด ํฌ๋ฉ”์ด์…˜์ด ๋  ์ˆ˜ ์žˆ๊ณ ,
10:14
they can be three-dimensional formations.
216
614497
2103
3์ฐจ์› ํฌ๋ฉ”์ด์…˜์ด ๋  ์ˆ˜ ๋„ ์žˆ์Šต๋‹ˆ๋‹ค.
์—ฌ๊ธฐ ๋ณด์‹œ๋Š” ๊ฒƒ ์ฒ˜๋Ÿผ,
10:17
As you can see here,
217
617331
1341
10:18
they collapse from a three-dimensional formation into planar formation.
218
618696
3687
3์ฐจ์› ํฌ๋ฉ”์ด์…˜์—์„œ ํ‰๋ฉด ํฌ๋ฉ”์ด์…˜์œผ๋กœ ํํŠธ๋Ÿฌ์ง‘๋‹ˆ๋‹ค.
๋˜, ์žฅ์• ๋ฌผ์„ ํ†ต๊ณผํ•˜๋ฉฐ ๋‚ ๊ธฐ ์œ„ํ•ด์„œ
10:22
And to fly through obstacles,
219
622407
1531
10:23
they can adapt the formations on the fly.
220
623962
2952
๋น„ํ–‰์ค‘์— ํฌ๋ฉ”์ด์…˜์— ์ ์‘ ํ•  ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค.
๋‹ค์‹œ ์ด ๋กœ๋ด‡๋“ค์ด ์•„์ฃผ ๊ฐ€๊น๊ฒŒ ๋‹ค๊ฐ€์˜ต๋‹ˆ๋‹ค.
10:28
So again, these robots come really close together.
221
628026
2463
10:30
As you can see in this figure-eight flight,
222
630513
2388
๊ทธ๋ฆผ์—์„œ ๋ณด์‹ค ์ˆ˜ ์žˆ๋“ฏ์ด,
10:32
they come within inches of each other.
223
632925
1984
๋กœ๋ด‡๋“ค์„ ์„œ๋กœ ๋ช‡ ์ธ์น˜์˜ ๊ฐ„๊ฒฉ์„ ๋‘๊ณ  ๋ถ™์„ ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค.
ํ”„๋กœํŽ ๋Ÿฌ ๋‚ ๊ฐœ์˜
10:35
And despite the aerodynamic interactions with these propeller blades,
224
635377
4317
๊ณต๊ธฐ์—ญํ•™์ ์ธ ์ƒํ˜ธ์ž‘์šฉ์—๋„ ๋ถˆ๊ตฌํ•˜๊ณ 
10:39
they're able to maintain stable flight.
225
639718
2148
์•ˆ์ •์ ์ธ ๋น„ํ–‰์„ ํ•  ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค.
10:41
(Applause)
226
641890
6896
(๋ฐ•์ˆ˜)
ํฌ๋ฉ”์ด์…˜์„ ์ด๋ฃจ๋ฉด์„œ ๋น„ํ–‰ํ•˜๋Š” ๋ฐฉ๋ฒ•์„ ์•Œ๊ฒŒ ๋˜๋ฉด
10:49
So once you know how to fly in formation,
227
649246
1999
ํ˜‘๋™ํ•ด์„œ ๋ฌผ๊ฑด์„ ๋“ค์–ด ์˜ฌ๋ฆด ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค.
10:51
you can actually pick up objects cooperatively.
228
651269
2216
์—ฌ๊ธฐ ๋ณด์‹ค ์ˆ˜ ์žˆ๋Š”๊ฒƒ ์ฒ˜๋Ÿผ,
10:53
So this just shows that we can double, triple, quadruple
229
653509
4577
๋กœ๋ด‡๋“ค์„ ํ•จ๊ป˜ ํŒ€์œผ๋กœ ๋งŒ๋“ค์–ด์„œ
๋กœ๋ด‡์˜ ํž˜์„ ๋‘ ๋ฐฐ, ์„ธ ๋ฐฐ,
10:58
the robots' strength,
230
658110
1158
10:59
by just getting them to team with neighbors, as you can see here.
231
659292
3077
๋„ค ๋ฐฐ ๋” ํฌ๊ฒŒ ๋งŒ๋“ค ์ˆ˜ ์žˆ๋‹ค๋Š” ๊ฒƒ์„ ๋ณด์—ฌ์ค๋‹ˆ๋‹ค.
์ด๋ ‡๊ฒŒ ํ•˜๋Š”๋ฐ ์žˆ์–ด ๋‹จ์ ์€
11:02
One of the disadvantages of doing that is, as you scale things up --
232
662393
4142
๋ฌผ๊ฑด์˜ ํฌ๊ธฐ๋ฅผ ๋Š˜๋ฆด์ˆ˜๋ก
11:06
so if you have lots of robots carrying the same thing,
233
666559
2564
๊ฐ™์€ ๋ฌผ๊ฑด์„ ์˜ฎ๊ธฐ๋Š” ๋งŽ์€ ๋กœ๋ด‡์ด ์žˆ๋‹ค๊ณ  ๊ฐ€์ •ํ•  ๋•Œ,
๊ธฐ๋ณธ์ ์œผ๋กœ ๊ด€์„ฑ์ด ์ฆ๊ฐ€ํ•˜๊ณ , ๋”ฐ๋ผ์„œ
11:09
you're essentially increasing the inertia,
234
669147
2720
11:11
and therefore you pay a price; they're not as agile.
235
671891
2619
๊ทธ๋งŒํผ ๋Œ“๊ฐ€๋ฅผ ์น˜๋ค„์•ผ ํ•˜๋Š”๋ฐ ๊ฒฐ๊ตญ ๋กœ๋ด‡์ด ๋ฏผ์ฒฉํ•ด์ง€์ง€ ์•Š๋Š”๋‹ค๋Š”๊ฑฐ์ฃ .
11:14
But you do gain in terms of payload-carrying capacity.
236
674978
3096
ํ•˜์ง€๋งŒ ์ ์žฌํ•˜์ค‘์˜ ์šฉ๋Ÿ‰์— ๋Œ€ํ•ด์„œ๋Š” ์ด์ ์„ ์–ป๊ฒŒ ๋ฉ๋‹ˆ๋‹ค.
๋ณด์—ฌ๋“œ๋ฆฌ๊ณ  ์‹ถ์€ ๋˜๋‹ค๋ฅธ ์‘์šฉ๋ถ„์•ผ๋Š”
11:18
Another application I want to show you -- again, this is in our lab.
237
678098
3215
-- ์—ฌ๊ธฐ๋Š” ์ €ํฌ ์—ฐ๊ตฌ์‹ค ์ž…๋‹ˆ๋‹ค. --
11:21
This is work done by Quentin Lindsey, who's a graduate student.
238
681337
3165
๋Œ€ํ•™์›์ƒ์ธ ์ฟ ์—”ํ‹ด ๋ฆฐ์ง€ ํ•™์ƒ์ด ์ž‘์—…ํ•œ ๊ฒƒ์ธ๋ฐ์š”.
๊ทธ์˜ ์•Œ๊ณ ๋ฆฌ์ฆ˜์€ ์ด๋Ÿฐ ๋กœ๋ด‡๋“ค์—๊ฒŒ
11:24
So his algorithm essentially tells these robots
239
684526
2620
๊ต๋Ÿ‰ ๊ตฌ์กฐ๋ฌผ์—์„œ
11:27
how to autonomously build cubic structures
240
687170
4072
์–ด๋–ป๊ฒŒ ์œก๋ฉด์ฒด ๊ตฌ์กฐ๋ฌผ์„ ๋งŒ๋“œ๋Š”์ง€
์•Œ๋ ค์ฃผ๊ณ  ์žˆ์Šต๋‹ˆ๋‹ค.
11:31
from truss-like elements.
241
691266
1970
๊ทธ๋ž˜์„œ ๊ทธ์˜ ์•Œ๊ณ ๋ฆฌ์ฆ˜์€
11:34
So his algorithm tells the robot what part to pick up,
242
694441
3760
์–ด๋–ค ๋ถ€๋ถ„์„ ๋“ค์–ด์˜ฌ๋ฆฌ๊ณ , ์–ธ์ œ, ์–ด๋””๋กœ ์˜ฎ๊ธฐ๋Š”์ง€๋ฅผ
๋กœ๋ด‡์—๊ฒŒ ๋งํ•ด์ค๋‹ˆ๋‹ค.
11:38
when, and where to place it.
243
698225
1825
์—ฌ๋Ÿฌ๋ถ„์€ ์ด ๋น„๋””์˜ค์—์„œ
11:40
So in this video you see --
244
700471
1483
11:41
and it's sped up 10, 14 times --
245
701978
2024
-- ์†๋„๋ฅผ 10, 14๋ฐฐ ์˜ฌ๋ฆฌ์ฃ  --
๋กœ๋ด‡๋“ค์— ์˜ํ•ด ๋งŒ๋“ค์–ด์ง„ 3๊ฐœ์˜ ์„œ๋กœ ๋‹ค๋ฅธ ๊ตฌ์กฐ๋ฌผ๋“ค์„ ๋ณด์‹ค ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค.
11:44
you see three different structures being built by these robots.
246
704026
3089
๋ชจ๋“ ๊ฒƒ์ด ์ž์œจ์ ์ž…๋‹ˆ๋‹ค.
11:47
And again, everything is autonomous,
247
707139
2026
์ฟ ์—”ํ‹ด ํ•™์ƒ์ด ํ•ด์•ผ ํ•˜๋Š” ๊ฒƒ์€
11:49
and all Quentin has to do
248
709189
1283
11:50
is to give them a blueprint of the design that he wants to build.
249
710496
3935
๋งŒ๋“ค๊ณ  ์‹ถ์€ ๋””์ž์ธ์˜ ์ฒญ์‚ฌ์ง„์„
๋กœ๋ด‡๋“ค์—๊ฒŒ ์•Œ๋ ค๋งŒ ์ฃผ๋Š” ๊ฒƒ์ž…๋‹ˆ๋‹ค.
11:56
So all these experiments you've seen thus far,
250
716925
2811
์ง€๊ธˆ๊นŒ์ง€ ๋ณด์…จ๋˜ ๋ชจ๋“  ์‹คํ—˜๋“ค๊ณผ,
11:59
all these demonstrations,
251
719760
1350
๋ชจ๋“  ์‹œ์—ฐ๋“ค์€
12:01
have been done with the help of motion-capture systems.
252
721134
3102
๋™์ž‘๊ฐ์ง€ ์‹œ์Šคํ…œ์˜ ๋„์›€์œผ๋กœ ์ด๋ฃจ์–ด์กŒ์Šต๋‹ˆ๋‹ค.
์—ฐ๊ตฌ์‹ค์„ ๋– ๋‚˜ ์‹ค์ œ ์„ธ์ƒ์œผ๋กœ ๋‚˜๊ฐ€๋ฉด
12:05
So what happens when you leave your lab,
253
725029
2507
๋ฌด์Šจ์ผ์ด ์ผ์–ด๋‚ ๊นŒ์š”?
12:07
and you go outside into the real world?
254
727560
2037
12:09
And what if there's no GPS?
255
729922
1706
GPS๊ฐ€ ์—†๋‹ค๋ฉด ์–ด๋–จ๊นŒ์š”?
12:12
So this robot is actually equipped with a camera,
256
732755
4337
๊ทธ๋ž˜์„œ ์ด๋Ÿฐ ๋กœ๋ด‡์€
์นด๋ฉ”๋ผ, ๋ ˆ์ด์ ธ H ํŒŒ์ธ๋”, ์Šค์บ๋„ˆ๊ฐ€
์žฅ์ฐฉ๋˜์–ด ์žˆ์Šต๋‹ˆ๋‹ค.
12:17
and a laser rangefinder, laser scanner.
257
737116
2373
์ด๊ฒƒ์€ ์ฃผ๋ณ€ํ™˜๊ฒฝ์˜ ์ง€๋„๋ฅผ ๋งŒ๋“ค๊ธฐ ์œ„ํ•ด
12:20
And it uses these sensors to build a map of the environment.
258
740140
3920
์ด๋Ÿฐ ์„ผ์„œ๋“ค์„ ์‚ฌ์šฉํ•ฉ๋‹ˆ๋‹ค.
์ง€๋„๋Š” ๋ช‡๊ฐ€์ง€ ๊ตฌ์กฐ๋ฌผ๋“ค๋กœ ๊ตฌ์„ฑ๋˜๋Š”๋ฐ์š”
12:24
What that map consists of are features --
259
744084
3286
-- ์ถœ์ž…๊ตฌ, ์ฐฝ๋ฌธ๋“ค,
12:27
like doorways, windows, people, furniture --
260
747394
3871
์‚ฌ๋žŒ๋“ค, ๊ฐ€๊ตฌ ๋“ฑ --
๊ทธ๋ฆฌ๊ณ  ๊ทธ๊ฒƒ๋“ค์ด ๊ตฌ์กฐ๋ฌผ์— ์—ฐ๊ด€ํ•ด
12:31
and it then figures out where its position is,
261
751289
2143
์–ด๋””์— ์œ„์น˜ํ•˜๋Š”์ง€๋ฅผ ๊ณ„์‚ฐํ•ฉ๋‹ˆ๋‹ค.
12:33
with respect to the features.
262
753456
1401
12:34
So there is no global coordinate system.
263
754881
2210
๊ทธ๋ž˜์„œ ์ „์—ญ์ ์ธ ์ขŒํ‘œ ์‹œ์Šคํ…œ์ด ์—†์Šต๋‹ˆ๋‹ค.
์ขŒํ‘œ ์‹œ์Šคํ…œ์€ ๋กœ๋ด‡์„ ๊ธฐ๋ฐ˜์œผ๋กœ ์ •์˜ ๋˜๋Š”๋ฐ์š”,
12:37
The coordinate system is defined based on the robot,
264
757115
2506
12:39
where it is and what it's looking at.
265
759645
2057
์–ด๋””์— ์žˆ๋Š”์ง€, ์–ด๋””๋ฅผ ๋ณด๊ณ  ์žˆ๋Š”์ง€๋ฅผ ํ™•์ธ ํ•ฉ๋‹ˆ๋‹ค.
12:42
And it navigates with respect to those features.
266
762575
2755
๊ทธ๋ฆฌ๊ณ  ๊ทธ๋Ÿฐ ๊ตฌ์กฐ๋ฌผ๋“ค์„ ๋”ฐ๋ผ์„œ ๋น„ํ–‰ํ•˜๊ฒŒ ๋ฉ๋‹ˆ๋‹ค.
ํ”„๋žญํฌ ์‰”๊ณผ
12:46
So I want to show you a clip
267
766316
1420
12:47
of algorithms developed by Frank Shen and Professor Nathan Michael,
268
767760
3952
๋„ค์ด์ฌ ๋งˆ์ดํด ๊ต์ˆ˜๊ฐ€ ๊ฐœ๋ฐœํ•œ
์•Œ๊ณ ๋ฆฌ์ฆ˜์˜ ๋™์˜์ƒ์„ ๋ณด๋“œ๋ฆฌ๊ณ  ์‹ถ์Šต๋‹ˆ๋‹ค.
12:51
that shows this robot entering a building for the very first time,
269
771736
3769
์ด ๋™์˜์ƒ์€ ๋งจ ์ฒ˜์Œ ๋กœ๋ด‡์ด ๋นŒ๋”ฉ์— ๋“ค์–ด๊ฐ€์„œ
12:55
and creating this map on the fly.
270
775529
2422
๋น„ํ–‰์ค‘์— ์ง€๋„๋ฅผ ์ž‘์„ฑํ•˜๋Š” ์žฅ๋ฉด์„ ๋ณด์—ฌ์ค๋‹ˆ๋‹ค.
12:58
So the robot then figures out what the features are,
271
778642
3158
๊ทธ ๋กœ๋ด‡์€ ๋ฌด์Šจ ๊ตฌ์กฐ๋ฌผ๋“ค์ด ์žˆ๋Š”์ง€๋ฅผ ์•Œ์•„๋ƒ…๋‹ˆ๋‹ค.
13:01
it builds the map,
272
781824
1151
๊ทธ๋ฆฌ๊ณ  ์ง€๋„๋ฅผ ๋งŒ๋“ญ๋‹ˆ๋‹ค.
13:02
it figures out where it is with respect to the features,
273
782999
2967
๋กœ๋ด‡์€ ๊ตฌ์กฐ๋ฌผ๋“ค์„ ๋น„๊ตํ•˜์—ฌ ์–ด๋””์— ์œ„์น˜ํ•˜๋Š”์ง€๋ฅผ ๊ณ„์‚ฐํ•˜๊ณ 
13:05
and then estimates its position 100 times a second,
274
785990
3948
์ดˆ๋‹น 100๋ฒˆ์ •๋„ ๋กœ๋ด‡์˜ ์œ„์น˜๋ฅผ ์ธก์ •ํ•ฉ๋‹ˆ๋‹ค.
์ฒ˜์Œ์— ์ œ๊ฐ€ ์„ค๋ช…๋“œ๋ฆฐ
13:09
allowing us to use the control algorithms that I described to you earlier.
275
789962
3818
์ œ์–ด ์•Œ๊ณ ๋ฆฌ์ฆ˜์„
์‚ฌ์šฉ ํ•  ์ˆ˜ ์žˆ๊ฒŒ ํ•œ ๊ฒƒ์ด์ฃ .
13:13
So this robot is actually being commanded remotely by Frank,
276
793804
4433
์ด ๋กœ๋ด‡์€ ํ”„๋žญํฌ์— ์˜ํ•ด
์›๊ฒฉ์œผ๋กœ ๋ช…๋ น์ด ๋‚ด๋ ค์ง„๊ฒƒ์ž…๋‹ˆ๋‹ค.
ํ•˜์ง€๋งŒ ๋กœ๋ด‡์€ ์–ด๋””๋กœ ๊ฐ€์•ผํ•˜๋Š”์ง€
13:18
but the robot can also figure out where to go on its own.
277
798261
3539
์ž์‹ ์ด ํ•ด๊ฒฐํ•ฉ๋‹ˆ๋‹ค.
์ œ๊ฐ€ ์ด ๋กœ๋ด‡์„ ๋นŒ๋”ฉ์•ˆ์œผ๋กœ ๋“ค์—ฌ ๋ณด๋ƒˆ๋‹ค๊ณ  ๊ฐ€์ •ํ•ด๋ณด์ฃ .
13:22
So suppose I were to send this into a building,
278
802125
2207
์ €๋Š” ์ด ๋นŒ๋”ฉ์ด ์–ด๋–ป๊ฒŒ ์ƒ๊ฒผ๋Š”์ง€ ๋ชจ๋ฆ…๋‹ˆ๋‹ค.
13:24
and I had no idea what this building looked like.
279
804356
2302
์ด ๋กœ๋ด‡์—๊ฒŒ ๋“ค์–ด๊ฐ€๋ผ๊ณ  ๋ช…๋ นํ•˜๊ณ ,
13:26
I can ask this robot to go in,
280
806682
1444
์ง€๋„๋ฅผ ๋งŒ๋“ค๋ผ๊ณ  ํ•˜๊ณ ,
13:28
create a map,
281
808150
1904
๋Œ์•„์™€์„œ ๋นŒ๋”ฉ์ด ์–ด๋–ค ๋ชจ์–‘์ธ์ง€ ๋งํ•ด๋‹ฌ๋ผ๊ณ  ํ•  ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค.
13:30
and then come back and tell me what the building looks like.
282
810078
2865
13:32
So here, the robot is not only solving the problem
283
812967
3168
์—ฌ๊ธฐ๋ณด๋ฉด, ๊ทธ ๋กœ๋ด‡๋“ค์€ ์ง€๋„์˜ A์ง€์ ์—์„œ
B์ง€์ ์œผ๋กœ ์–ด๋–ป๊ฒŒ ๊ฐ€๋Š”์ง€ ๋ฟ๋งŒ ์•„๋‹ˆ๋ผ,
13:36
of how to go from point A to point B in this map,
284
816159
2634
13:38
but it's figuring out what the best point B is at every time.
285
818817
4399
๋งค๋ฒˆ ์–ด๋–ค ์ง€์ ์ด ์ตœ์ ์˜ B์ง€์ ์ธ์ง€๋„
์•Œ์•„๋‚ผ ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค.
์ตœ์†Œํ•œ์˜ ์ •๋ณด๋ฅผ ๊ฐ€์ง€๊ณ  ํŠน์ • ์žฅ์†Œ๋ฅผ ์ฐพ๊ธฐ์œ„ํ•ด
13:43
So essentially it knows where to go
286
823240
2054
13:45
to look for places that have the least information,
287
825318
2785
์–ด๋””๋กœ ๊ฐ€์•ผํ•˜๋Š”์ง€ ์•Œ๊ณ  ์žˆ์Šต๋‹ˆ๋‹ค.
๋˜ํ•œ, ๋กœ๋ด‡์€ ์–ด๋–ป๊ฒŒ ์ง€๋„์— ๋ง๋ถ™์ด๋Š”์ง€๋ฅผ ์•Œ๋ ค์ค๋‹ˆ๋‹ค.
13:48
and that's how it populates this map.
288
828127
1968
13:50
So I want to leave you with one last application.
289
830944
2998
๋งˆ์ง€๋ง‰ ํ•œ๊ฐ€์ง€ ์‘์šฉ๋ถ„์•ผ๋ฅผ
๋ณด์—ฌ ๋“œ๋ฆฌ๊ฒ ์Šต๋‹ˆ๋‹ค.
13:54
And there are many applications of this technology.
290
834982
2524
์ด ๊ธฐ์ˆ ์—๋Š” ๋งŽ์€ ์‘์šฉ๋ถ„์•ผ๊ฐ€ ์žˆ์Šต๋‹ˆ๋‹ค.
13:57
I'm a professor, and we're passionate about education.
291
837823
2548
์ €๋Š” ๊ต์ˆ˜์ž…๋‹ˆ๋‹ค. ๊ทธ๋ฆฌ๊ณ  ๊ต์œก์— ๋Œ€ํ•ด ๋งค์šฐ ์—ด์ •์ ์ž…๋‹ˆ๋‹ค.
์ด์™€ ๊ฐ™์€ ๋กœ๋ด‡์€ ์ดˆ๋“ฑํ•™๊ต ๊ต์œก๊ณผ์ •์„
14:00
Robots like this can really change the way we do K-12 education.
292
840395
3936
๋ฐ”๊ฟ€ ์ˆ˜๋„ ์žˆ์Šต๋‹ˆ๋‹ค.
ํ•˜์ง€๋งŒ ์ €ํฌ๋Š” ๋กœ์Šค์•ค์ ค๋ ˆ์Šค์— ๊ฐ€๊นŒ์šด
14:04
But we're in Southern California,
293
844355
1825
๋‚จ๋ถ€ ์บ˜๋ฆฌํฌ๋‹ˆ์•„ ์ง€๋ฐฉ์— ์žˆ์Šต๋‹ˆ๋‹ค.
14:06
close to Los Angeles,
294
846204
1833
๊ทธ๋ž˜์„œ ์—”ํ„ฐํ…Œ์ธ๋จผํŠธ์— ๊ด€๋ จ๋œ ๊ฒƒ์œผ๋กœ
14:08
so I have to conclude with something focused on entertainment.
295
848061
3250
๊ฒฐ๋ก ์„ ๋งบ๊ณ  ์‹ถ๊ตฐ์š”.
๋ฎค์ง๋น„๋””์˜ค ํ•œํŽธ์œผ๋กœ ๊ฐ•์—ฐ์„ ๋งˆ์น˜๊ณ  ์‹ถ์Šต๋‹ˆ๋‹ค.
14:12
I want to conclude with a music video.
296
852407
1834
์ด ๋น„๋””์˜ค๋ฅผ ๋งŒ๋“ 
14:14
I want to introduce the creators, Alex and Daniel, who created this video.
297
854265
4714
์•Œ๋ ‰์Šค์™€ ๋‹ค๋‹ˆ์—˜์„ ์†Œ๊ฐœํ•ฉ๋‹ˆ๋‹ค.
(๋ฐ•์ˆ˜)
14:19
(Applause)
298
859003
6878
14:25
So before I play this video,
299
865905
1434
์ด ๋น„๋””์˜ค๋ฅผ ์‹คํ–‰ํ•˜๊ธฐ ์ „์—,
14:27
I want to tell you that they created it in the last three days,
300
867363
3197
ํฌ๋ฆฌ์Šค๋กœ๋ถ€ํ„ฐ ์ „ํ™”๋ฅผ ๋ฐ›๊ณ  3์ผ๋งŒ์—
14:30
after getting a call from Chris.
301
870584
1652
์ด ์ž‘ํ’ˆ์„ ๋งŒ๋“ค์—ˆ๋‹ค๋Š” ๊ฒƒ์„ ๋ง์”€ ๋“œ๋ฆฌ๊ณ  ์‹ถ์Šต๋‹ˆ๋‹ค.
14:32
And the robots that play in the video are completely autonomous.
302
872743
3817
๋น„๋””์˜ค ์ƒ์˜๋˜๋Š” ๋กœ๋ด‡๋“ค์€
์™„์ „ํžˆ ์ž์œจ์ ์œผ๋กœ ์›€์ง์ž…๋‹ˆ๋‹ค.
14:36
You will see nine robots play six different instruments.
303
876584
3502
6๊ฐœ์˜ ์„œ๋กœ ๋‹ค๋ฅธ ์•…๊ธฐ๋ฅผ ๋‹ค๋ฃจ๋Š” 9๊ฐœ์˜ ๋กœ๋ด‡๋“ค์„ ๋ณด์‹ค๊ฒ๋‹ˆ๋‹ค.
๋ฌผ๋ก  TED 2012๋ฅผ ์œ„ํ•ด์„œ ๋งŒ๋“ค์—ˆ์Šต๋‹ˆ๋‹ค.
14:40
And of course, it's made exclusively for TED 2012.
304
880697
2971
์ž ๋ณด์‹œ์ฃ .
14:44
Let's watch.
305
884533
1152
14:46
(Sound of air escaping from valve)
306
886824
4404
14:53
(Music)
307
893547
3336
14:56
(Whirring sound)
308
896907
5520
15:19
(Music)
309
919442
6629
(์Œ์•…)
(๋ฐ•์ˆ˜)
16:24
(Applause) (Cheers)
310
984029
5984
์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