Robots with "soul" | Guy Hoffman

1,554,560 views ใƒป 2014-01-17

TED


ืื ื ืœื—ืฅ ืคืขืžื™ื™ื ืขืœ ื”ื›ืชื•ื‘ื™ื•ืช ื‘ืื ื’ืœื™ืช ืœืžื˜ื” ื›ื“ื™ ืœื”ืคืขื™ืœ ืืช ื”ืกืจื˜ื•ืŸ.

ืžืชืจื’ื: Shir Ben Asher Kestin ืžื‘ืงืจ: Ido Dekkers
00:12
My job is to design, build and study robots that communicate with people.
0
12564
4643
ื”ืขื‘ื•ื“ื” ืฉืœื™ ื”ื™ื ืœืขืฆื‘, ืœื‘ื ื•ืช ื•ืœืœืžื•ื“
ืจื•ื‘ื•ื˜ื™ื ืฉื™ื›ื•ืœื™ื ืœืชืงืฉืจ ืขื ืื ืฉื™ื.
00:17
But this story doesn't start with robotics at all, it starts with animation.
1
17231
3619
ืื‘ืœ ื”ืกื™ืคื•ืจ ื”ื–ื” ืœื ืžืชื—ื™ืœ ื‘ืจื•ื‘ื•ื˜ื™ืงื” ื‘ื›ืœืœ,
ื”ื•ื ืžืชื—ื™ืœ ื‘ืื ื™ืžืฆื™ื”.
00:20
When I first saw Pixar's "Luxo Jr.,"
2
20874
2369
ื›ืฉืจืื™ืชื™ ืœืจืืฉื•ื ื” ืืช "ืœื•ืงืกื• ื’'ื•ื ื™ื•ืจ" ืฉืœ ืคื™ืงืกืืจ,
00:23
I was amazed by how much emotion they could put
3
23267
2706
ื ื“ื”ืžืชื™ ืžื›ืžื•ืช ื”ืจื’ืฉ
00:25
into something as trivial as a desk lamp.
4
25997
2828
ืฉื”ื ืžืกื•ื’ืœื™ื ืœื”ื›ื ื™ืก ื‘ืžืฉื”ื•
ื˜ืจื™ื•ื™ืืœื™ ื›ืžื• ืžื ื•ืจืช ืฉื•ืœื—ืŸ.
00:29
I mean, look at them -- at the end of this movie,
5
29477
2345
ื›ืœื•ืžืจ, ื”ืกืชื›ืœื• ืขืœื™ื”ื-- ื‘ืกื•ืฃ ื”ืกืจื˜,
00:31
you actually feel something for two pieces of furniture.
6
31846
2779
ืืชื” ื‘ืืžืช ืžืจื’ื™ืฉ ืžืฉื”ื• ื›ืœืคื™ ืฉืชื™ ื—ืชื™ื›ื•ืช ืจื™ื”ื•ื˜.
(ืฆื—ื•ืง)
00:34
(Laughter)
7
34649
1281
00:35
And I said, I have to learn how to do this.
8
35954
2028
ื•ืื ื™ ืืžืจืชื™, ืื ื™ ื—ื™ื™ื‘ ืœืœืžื•ื“ ืื™ืš ืœืขืฉื•ืช ืืช ื–ื”.
ืื– ืงื™ื‘ืœืชื™ ื”ื—ืœื˜ื” ืžืžืฉ ื’ืจื•ืขื” ืœืงืจื™ื™ืจื”.
00:38
So I made a really bad career decision.
9
38006
2227
00:40
(Laughter)
10
40257
1194
00:41
And that's what my mom was like when I did it.
11
41475
2193
ื•ื›ื›ื” ืืžื ืฉืœื™ ื ืจืืชื” ื›ืฉืขืฉื™ืชื™ ืืช ื–ื”.
00:43
(Laughter)
12
43692
2141
(ืฆื—ื•ืง)
00:45
I left a very cozy tech job in Israel at a nice software company
13
45857
3421
ืขื–ื‘ืชื™ ืขื‘ื•ื“ื” ื˜ื›ื ื•ืœื•ื’ื™ืช ืžืื•ื“ ื ื•ื—ื” ื‘ื™ืฉืจืืœ
ื‘ื—ื‘ืจืช ืชื•ื›ื ื” ื ื—ืžื“ื” ื•ืขื‘ืจืชื™ ืœื ื™ื• ื™ื•ืจืง
00:49
and I moved to New York to study animation.
14
49302
2292
ืœืœืžื•ื“ ืื ื™ืžืฆื™ื”.
00:51
And there I lived
15
51618
1151
ื•ืฉื ื—ื™ื™ืชื™
00:52
in a collapsing apartment building in Harlem with roommates.
16
52793
2818
ื‘ื‘ื ื™ื™ืŸ ื“ื™ืจื•ืช ืžืชืžื•ื˜ื˜ ื‘ื”ืืจืœื ืขื ืฉื•ืชืคื™ื.
00:55
I'm not using this phrase metaphorically --
17
55635
2024
ืื ื™ ืœื ืžืฉืชืžืฉ ื‘ืžืฉืคื˜ ื”ื–ื” ื‘ืื•ืคืŸ ืžื˜ืืคื•ืจื™,
00:57
the ceiling actually collapsed one day in our living room.
18
57683
2729
ื”ืชืงืจื” ื‘ืืžืช ื”ืชืžื•ื˜ื˜ื” ื™ื•ื ืื—ื“
ื‘ืกืœื•ืŸ ืฉืœื ื•.
01:00
Whenever they did news stories about building violations in New York,
19
60436
3242
ืชืžื™ื“ ื›ืฉืขืฉื• ื›ืชื‘ื•ืช ื‘ื—ื“ืฉื•ืช ืขืœ ืขื‘ื™ืจื•ืช ื‘ื ื™ื™ื” ื‘ื ื™ื• ื™ื•ืจืง,
ื”ื™ื• ืขื•ืฉื™ื ืืช ื”ื“ื™ื•ื•ื— ืžื•ืœ ื”ื‘ื ื™ื™ืŸ ืฉืœื ื•.
01:03
they would put the report in front of our building,
20
63702
2416
ื‘ืชื•ืจ ืกื•ื’ ืฉืœ ืชืคืื•ืจื” ืœื”ืจืื•ืช ื›ืžื” ื’ืจื•ืข ื”ืžืฆื‘.
01:06
as kind of, like, a backdrop to show how bad things are.
21
66142
2737
01:08
Anyway, during the day, I went to school
22
68903
1937
ื‘ื›ืœ ืžืงืจื”, ื‘ืžื”ืœืš ื”ื™ื•ื ื”ื™ื™ืชื™ ื”ื•ืœืš ืœืœื™ืžื•ื“ื™ื ื•ื‘ืœื™ืœื”
01:10
and at night I would sit and draw frame by frame of pencil animation.
23
70864
3427
ื”ื™ื™ืชื™ ื™ื•ืฉื‘ ื•ืžืฆื™ื™ืจ ืคืจื™ื™ื ืื—ืจื™ ืคืจื™ื™ื ืฉืœ ืื ื™ืžืฆื™ื” ื‘ืขื™ืคืจื•ืŸ.
01:14
And I learned two surprising lessons.
24
74315
2185
ื•ืœืžื“ืชื™ ืฉื ื™ ืœืงื—ื™ื ืžืคืชื™ืขื™ื--
01:16
One of them was that when you want to arouse emotions,
25
76524
4490
ืื—ื“ ืžื”ื ื”ื™ื”
ืฉื›ืฉืืชื” ืจื•ืฆื” ืœืขื•ืจืจ ืจื’ืฉื•ืช,
01:21
it doesn't matter so much how something looks;
26
81038
2175
ื–ื” ืœื ืžืฉื ื” ื›ืœ ื›ืš ืื™ืš ืžืฉื”ื• ื ืจืื”,
ื–ื” ื”ื›ืœ ื‘ืชื ื•ืขื”-- ื–ื” ื‘ืชื–ืžื•ืŸ
01:23
it's all in the motion, in the timing of how the thing moves.
27
83237
3198
ืฉืœ ืื™ืš ื”ื“ื‘ืจ ื ืข..
01:26
And the second was something one of our teachers told us.
28
86816
3190
ื•ื”ืฉื ื™ ื”ื™ื” ืžืฉื”ื• ืฉืื—ื“ ืžื”ืžื•ืจื™ื ืฉืœื ื• ืืžืจ ืœื ื•.
ืœืžืขืฉื” ื”ื•ื ื™ืฆืจ ืืช ื”ืกืžื•ืจ ื‘ืขื™ื“ืŸ ื”ืงืจื—.
01:30
He actually did the weasel in "Ice Age."
29
90030
2293
01:32
And he said, "As an animator, you're not a director -- you're an actor."
30
92662
3857
ื•ื”ื•ื ืืžืจ:
"ื›ืื ื™ืžื˜ื•ืจ ืืชื” ืœื ื‘ืžืื™, ืืชื” ืฉื—ืงืŸ."
01:36
So, if you want to find the right motion for a character,
31
96895
3023
ืื–, ืื ืืชื” ืจื•ืฆื” ืœืžืฆื•ื ืืช ื”ืชื ื•ืขื” ื”ื ื›ื•ื ื” ืขื‘ื•ืจ ื“ืžื•ืช,
01:39
don't think about it -- go use your body to find it.
32
99942
2445
ืืœ ืชื—ืฉื•ื‘ ืขืœ ื–ื”, ืœืš ืชืฉืชืžืฉ ื‘ื’ื•ืฃ ืฉืœืš ืขืœ ืžื ืช ืœืžืฆื•ื ืืช ื–ื”--
01:42
Stand in front of a mirror, act it out in front of a camera --
33
102411
2933
ืขืžื•ื“ ืžื•ืœ ืžืจืื”, ืฉื—ืง ืืช ื–ื”
ืžื•ืœ ืžืฆืœืžื”-- ืžื” ืฉืืชื” ืฆืจื™ืš.
01:45
whatever you need -- and then put it back in your character.
34
105368
2938
ื•ืื– ืชื›ื ื™ืก ืืช ื–ื” ื‘ื—ื–ืจื” ืœื“ืžื•ืช ืฉืœืš.
01:48
A year later I found myself at MIT in the Robotic Life Group.
35
108696
3212
ืฉื ื” ืœืื—ืจ ืžื›ืŸ ืžืฆืืชื™ ืืช ืขืฆืžื™ ื‘- MIT
ื‘ืงื‘ื•ืฆืช ื”ื—ื™ื™ื ื”ืจื•ื‘ื•ื˜ื™ื™ื, ื–ื• ื”ื™ื™ืชื” ืื—ืช ืžื”ืงื‘ื•ืฆื•ืช ื”ืจืืฉื•ื ื•ืช
01:51
It was one of the first groups researching the relationships
36
111932
2858
ืฉื—ืงืจื• ืืช ืžืขืจื›ื•ืช ื”ื™ื—ืกื™ื ืฉื‘ื™ืŸ ื‘ื ื™ ืื“ื ื•ืจื•ื‘ื•ื˜ื™ื.
01:54
between humans and robots.
37
114814
1314
ื•ืขื“ื™ื™ืŸ ื”ื™ื” ืœื™ ืืช ื”ื—ืœื•ื ื”ื–ื” ืœื™ืฆื•ืจ
01:56
And I still had this dream to make an actual, physical Luxo Jr. lamp.
38
116152
3956
ืžื ื•ืจืช ืœื•ืงืกื• ื’'ื•ื ื™ื•ืจ ืืžื™ืชื™, ืžื•ื—ืฉื™ืช.
02:00
But I found that robots didn't move at all in this engaging way
39
120132
2992
ืื‘ืœ ื’ื™ืœื™ืชื™ ืฉืจื•ื‘ื•ื˜ื™ื ืœื ื–ื–ื™ื ื‘ื›ืœืœ
ื‘ืื•ืคืŸ ืฉื•ื‘ื” ื”ืœื‘ ืฉื”ื™ื™ืชื™ ืจื’ื™ืœ ืืœื™ื•
ื‘ืœื™ืžื•ื“ื™ ื”ืื ื™ืžืฆื™ื” ืฉืœื™.
02:03
that I was used to from my animation studies.
40
123148
2175
ื‘ืžืงื•ื ื–ื”, ื”ื ื›ื•ืœื ื”ื™ื•--
02:05
Instead, they were all -- how should I put it --
41
125347
2464
ืื™ืš ืœื ืกื— ืืช ื–ื”, ื”ื ื›ื•ืœื ื”ื™ื• ืกื•ื’ ืฉืœ ืจื•ื‘ื•ื˜ื™ื™ื.
02:07
they were all kind of robotic.
42
127835
1454
02:09
(Laughter)
43
129313
1874
(ืฆื—ื•ืง)
02:11
And I thought, what if I took whatever I learned in animation school,
44
131211
3736
ื•ื—ืฉื‘ืชื™, ืžื” ืื ืื ื™ ืืงื— ืืช ืžื” ืฉืœืžื“ืชื™ ื‘ื‘ื™ืช ื”ืกืคืจ ืœืื ื™ืžืฆื™ื”,
02:14
and used that to design my robotic desk lamp.
45
134971
2603
ื•ืืฉืชืžืฉ ื‘ื–ื” ืขืœ ืžื ืช ืœืขืฆื‘ ืืช ื”ืžื ื•ืจื” ื”ืจื•ื‘ื•ื˜ื™ืช ืฉืœื™.
02:17
So I went and designed frame by frame
46
137598
2045
ืื– ื”ืœื›ืชื™ ื•ืขื™ืฆื‘ืชื™ ืคืจื™ื™ื ืื—ืจื™ ืคืจื™ื™ื
02:19
to try to make this robot as graceful and engaging as possible.
47
139667
3960
ืœื ืกื•ืช ืœื™ืฆื•ืจ ืืช ื”ืจื•ื‘ื•ื˜ ื”ื–ื”
ืœื—ื™ื ื ื™ ื•ืฉื•ื‘ื” ืœื‘ ื›ื›ืœ ื”ื ื™ืชืŸ.
02:24
And here when you see the robot interacting with me on a desktop --
48
144199
3709
ื•ื›ืืŸ ื›ืฉืืชื ืจื•ืื™ื ืืช ื”ืจื•ื‘ื•ื˜ ืžืชืงืฉืจ ืื™ืชื™
ืขืœ ืฉื•ืœื—ืŸ.
02:27
and I'm actually redesigning the robot,
49
147932
2036
ื•ืื ื™ ืœืžืขืฉื” ืžืขืฆื‘ ืžื—ื“ืฉ ืืช ื”ืจื•ื‘ื•ื˜ ื›ืš,
02:29
so, unbeknownst to itself,
50
149992
1804
ืžื‘ืœื™ ืฉื™ื”ื™ื” ื™ื“ื•ืข ืœืขืฆืžื•,
02:31
it's kind of digging its own grave by helping me.
51
151820
2575
ื”ื•ื ื‘ืขืจืš ื—ื•ืคืจ ืืช ื”ืงื‘ืจ ืฉืœ ืขืฆืžื• ืขืœ ื™ื“ื™ ื›ืš ืฉื”ื•ื ืขื•ื–ืจ ืœื™.
02:34
(Laughter)
52
154419
2006
(ืฆื—ื•ืง)
02:36
I wanted it to be less of a mechanical structure giving me light,
53
156449
3239
ืจืฆื™ืชื™ ืฉื–ื” ื™ื”ื™ื” ืคื—ื•ืช ืžื‘ื ื” ืžื›ื ื™
ืฉื ื•ืชืŸ ืœื™ ืื•ืจ,
02:39
and more of a helpful, kind of quiet apprentice
54
159712
3018
ื•ื™ื•ืชืจ ืฉื•ืœื™ื” ืฉื™ืžื•ืฉื™ ื•ืฉืงื˜
02:42
that's always there when you need it and doesn't really interfere.
55
162754
3169
ืฉืชืžื™ื“ ื ืžืฆื ืฉื ื›ืฉืืชื” ื–ืงื•ืง ืœื• ื•ืœื ืžืžืฉ ืžืชืขืจื‘.
02:45
And when, for example, I'm looking for a battery that I can't find,
56
165947
3335
ื•ื›ืืฉืจ, ืœื“ื•ื’ืžื”, ืื ื™ ืžื—ืคืฉ ืกื•ืœืœื”
ืฉืื ื™ ืœื ืžืฆืœื™ื— ืœืžืฆื•ื,
02:49
in a subtle way, it'll show me where the battery is.
57
169306
2947
ื‘ื“ืจืš ืขื“ื™ื ื”, ื”ื•ื ืžืจืื” ืœื™ ืื™ืคื” ื”ืกื•ืœืœื”.
02:53
So you can see my confusion here.
58
173872
1833
ืื– ืืชื ื™ื›ื•ืœื™ื ืœื”ื‘ื™ืŸ ืืช ื”ื‘ืœื‘ื•ืœ ืฉืœื™ ื›ืืŸ.
02:56
I'm not an actor.
59
176442
1212
ืื ื™ ืœื ืฉื—ืงืŸ.
03:00
And I want you to notice how the same mechanical structure
60
180585
3033
ื•ืื ื™ ืจื•ืฆื” ืฉืชื‘ื—ื™ื ื• ืื™ืš
ืื•ืชื• ืžื‘ื ื” ืžื›ื ื™ ื™ื›ื•ืœ ื‘ื ืงื•ื“ื” ืื—ืช,
03:03
can, at one point, just by the way it moves,
61
183642
2087
ืจืง ืขืœ ื™ื“ื™ ื”ื“ืจืš ืฉื‘ื” ื”ื•ื ื–ื–, ืœื”ื™ืจืื•ืช ืขื“ื™ืŸ ื•ืื›ืคืชื™--
03:05
seem gentle and caring and in the other case,
62
185753
2144
03:07
seem violent and confrontational.
63
187921
2371
ื•ื‘ืžืงืจื” ืื—ืจ, ืœื”ื™ืจืื•ืช ืืœื™ื ื•ืขื•ื™ื™ืŸ.
03:10
And it's the same structure, just the motion is different.
64
190316
2847
ื•ื–ื”ื• ืื•ืชื• ืžื‘ื ื”, ืจืง ื”ืชื ื•ืขื” ืฉื•ื ื”.
03:19
Actor: "You want to know something? Well, you want to know something?
65
199419
4572
ืฉื—ืงืŸ: "ืืชื” ืจื•ืฆื” ืœื“ืขืช ืžืฉื”ื•? ื•ื‘ื›ืŸ, ืืชื” ืจื•ืฆื” ืœื“ืขืช ืžืฉื”ื•?"
03:24
He was already dead!
66
204015
1889
"ื”ื•ื ื›ื‘ืจ ื”ื™ื” ืžืช"
03:25
Just laying there, eyes glazed over!"
67
205928
3902
"ืคืฉื•ื˜ ืฉื•ื›ื‘ ืœื• ื›ื›ื”, ืขื ืขื™ื ื™ื™ื ืžื–ื•ื’ื’ื•ืช!"
03:29
(Laughter)
68
209854
1021
(ืฆื—ื•ืง)
03:30
But, moving in a graceful way is just one building block
69
210899
3056
ืื‘ืœ, ืœื ื•ืข ื‘ืฆื•ืจื” ื—ื™ื ื ื™ืช ื–ื” ืจืง ืื—ืช ืžืื‘ื ื™ ื”ื‘ื ื™ื™ืŸ ืฉืœ ื›ืœ ื”ืžื‘ื ื” ื”ื–ื”
03:33
of this whole structure called human-robot interaction.
70
213979
2595
ืฉื ืงืจื ืื™ื ื˜ืจืืงืฆื™ื™ืช ืื“ื-ืจื•ื‘ื•ื˜.
03:36
I was, at the time, doing my PhD, I was working on human-robot teamwork,
71
216598
3453
ื‘ืื•ืชื• ื–ืžืŸ ืขืฉื™ืชื™ ืืช ื”ื“ื•ืงื˜ื•ืจื˜ ืฉืœื™,
ืขื‘ื“ืชื™ ืขืœ ืขื‘ื•ื“ืช ืฆื•ื•ืช ืฉืœ ืจื•ื‘ื•ื˜ื™ื ื•ื‘ื ื™ ืื“ื;
ืฆื•ื•ืชื™ื ืฉืœ ื‘ื ื™ ืื“ื ื•ืจื•ื‘ื•ื˜ื™ื ืฉืขื•ื‘ื“ื™ื ื™ื—ื“.
03:40
teams of humans and robots working together.
72
220075
2062
03:42
I was studying the engineering,
73
222161
1635
ืœืžื“ืชื™ ืืช ื”ื”ื ื“ืกื”,
03:43
the psychology, the philosophy of teamwork,
74
223820
2532
ื”ืคืกื™ื›ื•ืœื•ื’ื™ื”, ื”ืคื™ืœื•ืกื•ืคื™ื” ืฉืœ ืขื‘ื•ื“ืช ืฆื•ื•ืช.
03:46
and at the same time,
75
226376
1151
ื•ื‘ืื•ืชื• ื–ืžืŸ ืžืฆืืชื™ ืืช ืขืฆืžื™
03:47
I found myself in my own kind of teamwork situation,
76
227551
2494
ื‘ืžืขื™ืŸ ืกื™ื˜ื•ืืฆื™ื” ืฉืœ ืขื‘ื•ื“ืช ืฆื•ื•ืช
ืขื ื—ื‘ืจ ื˜ื•ื‘ ืฉืœื™ ืฉืœืžืขืฉื” ื ืžืฆื ืคื”.
03:50
with a good friend of mine, who's actually here.
77
230069
2252
ื•ื‘ืกื™ื˜ื•ืืฆื™ื” ื”ื–ื• ืื ื• ื™ื›ื•ืœื™ื ื‘ืงืœื•ืช ืœื“ืžื™ื™ืŸ ืจื•ื‘ื•ื˜ื™ื
03:52
And in that situation, we can easily imagine robots
78
232345
2404
03:54
in the near future being there with us.
79
234773
1900
ื‘ืขืชื™ื“ ื”ืงืจื•ื‘ ืฉื ืžืฆืื™ื ื›ืืŸ ืื™ืชื ื•.
03:56
It was after a Passover Seder.
80
236697
1690
ื–ื” ื”ื™ื” ืœืื—ืจ ืœื™ืœ ื”ืกื“ืจ,
03:58
We were folding up a lot of folding chairs,
81
238411
2055
ืงื™ืคืœื ื• ื”ืจื‘ื” ื›ื™ืกืื•ืช ืžืชืงืคืœื™ื,
04:00
and I was amazed at how quickly we found our own rhythm.
82
240490
2658
ื•ื ื“ื”ืžืชื™ ืžื”ืžื”ื™ืจื•ืช ืฉื‘ื” ืžืฆืื ื• ืืช ื”ืงืฆื‘ ืฉืœื ื•.
ื›ืœ ืื—ื“ ืขืฉื” ืืช ื”ื—ืœืง ืฉืœื•.
04:03
Everybody did their own part, we didn't have to divide our tasks.
83
243172
3110
ืœื ื”ื™ื™ื ื• ืฆืจื™ื›ื™ื ืœื—ืœืง ื‘ื™ื ื™ื ื• ืืช ื”ืชืคืงื™ื“ื™ื.
04:06
We didn't have to communicate verbally about this --
84
246306
2446
ืœื ื”ื™ื™ื ื• ืฆืจื™ื›ื™ื ืœืชืงืฉืจ ืžื™ืœื•ืœื™ืช ืœื’ื‘ื™ ื–ื”.
04:08
it all just happened.
85
248776
1189
ื–ื” ืคืฉื•ื˜ ืงืจื”.
04:09
And I thought, humans and robots don't look at all like this.
86
249989
2880
ื•ืื ื™ ื—ืฉื‘ืชื™,
ื‘ื ื™ ืื“ื ื•ืจื•ื‘ื•ื˜ื™ื ื‘ื›ืœืœ ืœื ื ืจืื™ื ื›ื›ื”.
04:12
When humans and robots interact, it's much more like a chess game:
87
252893
3131
ื›ืฉื‘ื ื™ ืื“ื ื•ืจื•ื‘ื•ื˜ื™ื ืžืชืงืฉืจื™ื ื–ื” ืขื ื–ื”,
ื–ื” ื”ืจื‘ื” ื™ื•ืชืจ ื“ื•ืžื” ืœืžืฉื—ืง ืฉื—ืžื˜.
ื”ืื“ื ืขื•ืฉื” ื“ื‘ืจ ืžืกื•ื™ื™ื,
04:16
the human does a thing, the robot analyzes whatever the human did,
88
256048
3121
ื”ืจื•ื‘ื•ื˜ ืžื ืชื— ืืช ืžื” ืฉื”ืื“ื ืขืฉื”,
ื•ืื– ื”ืจื•ื‘ื•ื˜ ืžื—ืœื™ื˜ ืžื” ืœืขืฉื•ืช ื”ืœืื”,
04:19
the robot decides what to do next, plans it and does it.
89
259193
2643
ืžืชื›ื ืŸ ื•ืขื•ืฉื” ื–ืืช.
04:21
Then the human waits, until it's their turn again.
90
261860
2355
ื•ืื– ื”ืื“ื ืžื—ื›ื”, ืขื“ ืฉืชื•ืจื• ื™ื’ื™ืข ืฉื•ื‘.
ืื–, ื–ื” ื”ืจื‘ื” ื™ื•ืชืจ ื“ื•ืžื” ืœืžืฉื—ืง ืฉื—ืžื˜
04:24
So it's much more like a chess game, and that makes sense,
91
264239
2731
ื•ื–ื” ื”ื’ื™ื•ื ื™ ื‘ื’ืœืœ ืฉืฉื—ืžื˜ ื–ื” ื ื”ื“ืจ
04:26
because chess is great for mathematicians and computer scientists.
92
266994
3111
ืขื‘ื•ืจ ืžืชืžื˜ื™ืงืื™ื ื•ืžื“ืขื ื™ ืžื—ืฉื‘.
ื–ื” ื”ื›ืœ ืงืฉื•ืจ ืœื ื™ืชื•ื— ืžื™ื“ืข,
04:30
It's all about information, analysis, decision-making and planning.
93
270129
3428
ืงื‘ืœืช ื”ื—ืœื˜ื•ืช ื•ืชื›ื ื•ืŸ.
04:33
But I wanted my robot to be less of a chess player,
94
273581
3730
ืื‘ืœ ืจืฆื™ืชื™ ืฉื”ืจื•ื‘ื•ื˜ ืฉืœื™ ื™ื”ื™ื” ืคื—ื•ืช ืฉื—ืงืŸ ืฉื—ืžื˜,
04:37
and more like a doer
95
277335
1793
ื•ื™ื•ืชืจ ื›ืžื• ืžื™ืฉื”ื• ืฉืขื•ืฉื”
04:39
that just clicks and works together.
96
279152
2001
ืฉืคืฉื•ื˜ ืžืชื—ื‘ืจ ื•ืขื•ื‘ื“ ื™ื—ื“.
04:41
So I made my second horrible career choice:
97
281177
3369
ืื– ืงื™ื‘ืœืชื™ ืืช ื”ื”ื—ืœื˜ื” ื”ืฉื ื™ื” ื”ื ื•ืจืื™ืช ืœืงืจื™ื™ืจื” ืฉืœื™:
04:44
I decided to study acting for a semester.
98
284570
2553
ื”ื—ืœื˜ืชื™ ืœืœืžื•ื“ ืžืฉื—ืง ื‘ืžืฉืš ืกืžืกื˜ืจ.
04:47
I took off from the PhD, I went to acting classes.
99
287147
2919
ื”ืกืชืœืงืชื™ ืžืœื™ืžื•ื“ื™ ื“ื•ืงื˜ื•ืจื˜ ื•ื”ืœื›ืชื™ ืœืฉื™ืขื•ืจื™ ืžืฉื—ืง.
04:50
I actually participated in a play --
100
290090
2038
ืœืžืขืฉื” ื”ืฉืชืชืคืชื™ ื‘ืžื—ื–ื”,
04:52
I hope thereโ€™s no video of that around still.
101
292152
2302
ืื ื™ ืžืงื•ื•ื” ืฉืื™ืŸ ื•ื™ื“ืื• ืฉืœ ื–ื” ืžืกืชื•ื‘ื‘ ืขื“ื™ื™ืŸ.
04:54
(Laughter)
102
294478
1038
04:55
And I got every book I could find about acting,
103
295540
2294
ื•ื”ืฉื’ืชื™ ื›ืœ ืกืคืจ ืฉื™ื›ื•ืœืชื™ ืœืžืฆื•ื ืขืœ ืžืฉื—ืง,
04:57
including one from the 19th century that I got from the library.
104
297858
3129
ื›ื•ืœืœ ืื—ื“ ืžื”ืžืื” ื”-19
ืฉืœืงื—ืชื™ ืžื”ืกืคืจื™ื™ื”.
05:01
And I was really amazed, because my name was the second name on the list --
105
301011
3613
ื•ื ื“ื”ืžืชื™ ืœื’ืžืจื™ ื›ื™ื•ื•ืŸ ืฉื”ืฉื ืฉืœื™ ื”ื™ื” ื”ืฉื ื™ ื‘ืจืฉื™ืžื”--
05:04
the previous name was in 1889.
106
304648
2000
ื”ืฉื ื”ืงื•ื“ื ื”ื™ื” ื‘-1889. (ืฆื—ื•ืง)
05:06
(Laughter)
107
306672
1015
05:07
And this book was kind of waiting for 100 years
108
307711
2249
ื•ื”ืกืคืจ ื”ื–ื” ื›ืื™ืœื• ื—ื™ื›ื” 100 ืฉื ื™ื
05:09
to be rediscovered for robotics.
109
309984
2332
ืขืœ ืžื ืช ืœื”ืชื’ืœื•ืช ืžื—ื“ืฉ ืขื‘ื•ืจ ืจื•ื‘ื•ื˜ื™ืงื”.
05:12
And this book shows actors
110
312340
1576
ื•ื”ืกืคืจ ื”ื–ื” ืžืจืื” ืœืฉื—ืงื ื™ื
05:13
how to move every muscle in the body
111
313940
2238
ืื™ืš ืœื”ื–ื™ื– ื›ืœ ืฉืจื™ืจ ื‘ื’ื•ืฃ
05:16
to match every kind of emotion that they want to express.
112
316202
2716
ื›ื“ื™ ืœื”ืชืื™ื ืœื›ืœ ืกื•ื’ ืฉืœ ืจื’ืฉ ืฉื”ื ืจื•ืฆื™ื ืœื”ื‘ื™ืข.
05:18
But the real revelation was when I learned about method acting.
113
318942
3026
ืื‘ืœ ื”ืชื’ืœื™ืช ื”ืืžื™ืชื™ืช ื”ื™ื™ืชื”
ื›ืฉืœืžื“ืชื™ ืขืœ ืžืฉื—ืง ืžืชื•ื“ื™.
05:21
It became very popular in the 20th century.
114
321992
2253
ื–ื” ื ื”ื™ื” ืžืื•ื“ ืคื•ืคื•ืœืจื™ ื‘ืžืื” ื”-20.
05:24
And method acting said
115
324269
1151
ื•ืžืฉื—ืง ืžืชื•ื“ื™ ืืžืจ, ืืชื” ืœื ืฆืจื™ืš ืœืชื›ื ืŸ ื›ืœ ืฉืจื™ืจ ื‘ื’ื•ืฃ ืฉืœืš.
05:25
you don't have to plan every muscle in your body;
116
325444
2300
ื‘ืžืงื•ื ื–ืืช, ืืชื” ืฆืจื™ืš ืœื”ืฉืชืžืฉ ื‘ื’ื•ืฃ ืฉืœืš ืขืœ ืžื ืช ืœืžืฆื•ื ืืช ื”ืชื ื•ืขื” ื”ื ื›ื•ื ื”.
05:27
instead, you have to use your body to find the right movement.
117
327768
2945
ืืชื” ืฆืจื™ืš ืœื”ืฉืชืžืฉ ื‘ื–ื›ืจื•ืŸ ื”ื—ื•ืฉื™ ืฉืœืš
05:30
You have to use your sense memory to reconstruct the emotions
118
330737
2880
ืขืœ ืžื ืช ืœื‘ื ื•ืช ืžื—ื“ืฉ ืืช ื”ืจื’ืฉื•ืช ื•ื‘ืขืฆื
05:33
and kind of think with your body to find the right expression --
119
333641
3038
ืœื—ืฉื•ื‘ ื‘ืขื–ืจืช ื”ื’ื•ืฃ ืฉืœืš ืœืฉื ืžืฆื™ืืช ื”ื”ื‘ืขื” ื”ื ื›ื•ื ื”.
05:36
improvise, play off your scene partner.
120
336703
1875
ืœืืœืชืจ, ืœืฉื—ืง ื™ื—ื“ ืขื ื”ืฉื•ืชืฃ ืฉืœืš ืœืกืฆื™ื ื”.
05:38
And this came at the same time
121
338602
1445
ื•ื–ื” ื”ื’ื™ืข ื‘ืื•ืชื• ื–ืžืŸ ืฉื‘ื• ืงืจืืชื™ ืขืœ ื”ื˜ืจื ื“ ื”ื–ื”
05:40
as I was reading about this trend in cognitive psychology,
122
340071
3106
ื‘ืคืกื™ื›ื•ืœื•ื’ื™ื” ืงื•ื’ื ื™ื˜ื™ื‘ื™ืช ืฉื ืงืจื ืชื•ื“ืขื” ื ื˜ื•ืขืช ื’ื•ืฃ.
05:43
called embodied cognition, which also talks about the same ideas.
123
343201
3252
ืฉื’ื ื”ื•ื ืžื“ื‘ืจ ืขืœ ืื•ืชื ืจืขื™ื•ื ื•ืช--
05:46
We use our bodies to think;
124
346477
1515
ืื ื• ืžืฉืชืžืฉื™ื ื‘ื’ื•ืคื ื• ืขืœ ืžื ืช ืœื—ืฉื•ื‘,
ืื ื—ื ื• ืœื ื—ื•ืฉื‘ื™ื ืจืง ืขื ืžื•ื—ื ื• ื•ืžืฉืชืžืฉื™ื ื‘ื’ื•ืคื ื• ืขืœ ืžื ืช ืœื–ื•ื–,
05:48
we don't just think with our brains and use our bodies to move,
125
348016
2957
05:50
but our bodies feed back into our brain
126
350997
1918
ืืœื ืฉื’ื ื”ื’ื•ืฃ ืฉืœื ื• ื ื•ืชืŸ ืžืฉื•ื‘ ืœื’ื•ืฃ
05:52
to generate the way that we behave.
127
352939
2165
ื›ื“ื™ ืœื™ื™ืฆืจ ืืช ื”ื“ืจืš ืฉื‘ื ืื ื• ืžืชื ื”ื’ื™ื.
ื•ื–ื” ื”ื™ื” ื›ืžื• ืžื›ืช ื‘ืจืง.
05:55
And it was like a lightning bolt.
128
355128
1596
05:56
I went back to my office,
129
356748
1214
ื—ื–ืจืชื™ ืœืžืฉืจื“ ืฉืœื™.
05:57
I wrote this paper, which I never really published,
130
357986
2405
ื›ืชื‘ืชื™ ืืช ื”ืžืืžืจ ื”ื–ื”-- ืฉืืฃ ืคืขื ืœื ื‘ืืžืช ืคืจืกืžืชื™
06:00
called "Acting Lessons for Artificial Intelligence."
131
360415
2532
ืฉื ืงืจื "ืฉื™ืขื•ืจื™ ืžืฉื—ืง ืขื‘ื•ืจ ืื™ื ื˜ืœื™ื’ื ืฆื™ื” ืžืœืื›ื•ืชื™ืช"
06:02
And I even took another month
132
362971
1381
ื•ืืคื™ืœื• ืœืงื—ืชื™ ื—ื•ื“ืฉ ื ื•ืกืฃ
06:04
to do what was then the first theater play
133
364376
2444
ืœืขืฉื•ืช ืืช ืžื” ืฉื”ื™ื” ืื– ืžื—ื–ื” ื”ืชืื˜ืจื•ืŸ ื”ืจืืฉื•ืŸ
06:06
with a human and a robot acting together.
134
366844
1953
ืขื ืื“ื ื•ืจื•ื‘ื•ื˜ ืฉืžืฉื—ืงื™ื ื™ื—ื“.
06:08
That's what you saw before with the actors.
135
368821
2393
ื–ื” ืžื” ืฉืจืื™ืชื ืงื•ื“ื ืœื›ืŸ ืขื ื”ืฉื—ืงื ื™ื.
06:12
And I thought:
136
372564
1151
ื•ืื ื™ ื—ืฉื‘ืชื™:
06:13
How can we make an artificial intelligence model --
137
373739
2937
ืื™ืš ืื ื—ื ื• ื™ื›ื•ืœื™ื ืœื™ืฆื•ืจ ืžื•ื“ืœ ืฉืœ ืื™ื ื˜ืœื™ื’ื ืฆื™ื” ืžืœืื›ื•ืชื™ืช--
06:16
a computer, computational model --
138
376700
1983
ืžื—ืฉื‘, ืžื•ื“ืœ ืžืžื•ื—ืฉื‘--
06:18
that will model some of these ideas of improvisation,
139
378707
2507
ืฉื™ื—ืงื” ื›ืžื” ืžื”ืจืขื™ื•ื ื•ืช ื”ืืœื• ืฉืœ ืืœืชื•ืจ,
ืฉืœ ืœืงื™ื—ืช ืกื™ื›ื•ื ื™ื, ืฉืœ ืœืงื™ื—ืช ืฆ'ืื ืกื™ื,
06:21
of taking risks, of taking chances,
140
381238
1830
ืืคื™ืœื• ืฉืœ ืขืฉื™ื™ืช ื˜ืขื•ื™ื•ืช.
06:23
even of making mistakes?
141
383092
1527
06:24
Maybe it can make for better robotic teammates.
142
384643
2613
ืื•ืœื™ ื–ื” ื™ื›ื•ืœ ืœื”ื•ื‘ื™ืœ ืœื—ื‘ืจื™ ืฆื•ื•ืช ืจื•ื‘ื•ื˜ื™ื™ื ื˜ื•ื‘ื™ื ื™ื•ืชืจ.
06:27
So I worked for quite a long time on these models
143
387280
2604
ืื– ืขื‘ื“ืชื™ ื‘ืžืฉืš ื“ื™ ื”ืจื‘ื” ื–ืžืŸ ืขืœ ื”ืžื•ื“ืœื™ื ื”ืœืœื•
06:29
and I implemented them on a number of robots.
144
389908
2395
ื•ื™ื™ืฉืžืชื™ ืื•ืชื ืขืœ ืžืกืคืจ ืจื•ื‘ื•ื˜ื™ื.
06:32
Here you can see a very early example
145
392327
2301
ื›ืืŸ ืืชื ื™ื›ื•ืœื™ื ืœืจืื•ืช ื“ื•ื’ืžื” ืžืื•ื“ ืžื•ืงื“ืžืช
06:34
with the robots trying to use this embodied artificial intelligence
146
394652
3579
ืฉื‘ื” ื”ืจื•ื‘ื•ื˜ื™ื ืžื ืกื™ื ืœื”ืฉืชืžืฉ ื‘ืื™ื ื˜ืœื™ื’ื ื™ื” ื”ืžืœืื›ื•ืชื™ืช ื ื˜ื•ืขืช ื”ื’ื•ืฃ,
ืขืœ ืžื ืช ืœื ืกื•ืช ืœื”ืชืื™ื ืขืฆืžื ืœืชื ื•ืขื•ืช ืฉืœื™ ืขื“ ื›ืžื” ืฉื ื™ืชืŸ,
06:38
to try to match my movements as closely as possible.
147
398255
2463
06:40
It's sort of like a game.
148
400742
1463
ืงืฆืช ื›ืžื• ืžืฉื—ืง,
06:42
Let's look at it.
149
402530
1209
ื‘ื•ืื• ื ื‘ื™ื˜ ื‘ื–ื”.
06:47
You can see when I psych it out, it gets fooled.
150
407652
3539
ืืชื ื™ื›ื•ืœื™ื ืœืจืื•ืช ืฉื›ืฉืื ื™ ืžื ืกื” ืœืฉื’ืข ืื•ืชื•, ื”ื•ื ื ื•ืคืœ ื‘ืคื—.
06:51
And it's a little bit like what you might see actors do
151
411698
2595
ื•ื–ื” ืงืฆืช ื›ืžื• ืžื” ืฉืืคืฉืจ ืœืจืื•ืช ืฉืฉื—ืงื ื™ื ืขื•ืฉื™ื
ื›ืฉื”ื ืžื ืกื™ื ืœื”ื™ื•ืช ื“ืžื•ื™ื•ืช ืžืจืื” ืื—ื“ ืฉืœ ื”ืฉื ื™
06:54
when they try to mirror each other
152
414317
1634
06:55
to find the right synchrony between them.
153
415975
2365
ืขืœ ืžื ืช ืœืžืฆื•ื ืืช ื”ืกื™ื ื›ืจื•ื ื™ื–ืฆื™ื” ื”ื ื›ื•ื ื” ื‘ื™ื ื™ื”ื.
06:58
And then, I did another experiment,
154
418364
1832
ื•ืื–, ืขืฉื™ืชื™ ืขื•ื“ ื ื™ืกื•ื™,
07:00
and I got people off the street to use the robotic desk lamp,
155
420220
4059
ื•ืœืงื—ืชื™ ืื ืฉื™ื ืžื”ืจื—ื•ื‘ ืฉื™ืฉืชืžืฉื• ื‘ืžื ื•ืจืช ื”ืฉื•ืœื—ืŸ ื”ืจื•ื‘ื•ื˜ื™ืช ื”ื–ืืช,
07:04
and try out this idea of embodied artificial intelligence.
156
424303
3102
ื•ื™ื‘ื“ืงื• ืืช ื”ืจืขื™ื•ืŸ ืฉืœ ืื™ื ื˜ืœื™ื’ื ืฆื™ื” ืžืœืื›ื•ืชื™ืช ื ื˜ื•ืขืช ื’ื•ืฃ.
07:07
So, I actually used two kinds of brains for the same robot.
157
427921
4598
ืื–, ืœืžืขืฉื” ื”ืฉืชืžืฉืชื™ ื‘ืฉื ื™ ืกื•ื’ื™ื ืฉืœ ืžื•ื— ืขื‘ื•ืจ ืื•ืชื• ืจื•ื‘ื•ื˜.
07:12
The robot is the same lamp that you saw,
158
432543
1952
ื”ืจื•ื‘ื•ื˜ ื”ื•ื ืื•ืชื” ืžื ื•ืจื” ืฉืจืื™ืชื,
ื•ืฉืžืชื™ ื‘ืชื•ื›ื• ืฉื ื™ ืžื•ื—ื•ืช.
07:14
and I put two brains in it.
159
434519
1339
07:15
For one half of the people,
160
435882
1727
ืขื‘ื•ืจ ื—ืฆื™ ืžื”ืื ืฉื™ื,
07:17
I put in a brain that's kind of the traditional,
161
437633
2936
ืฉืžืชื™ ืžื•ื— ื“ื™ ืžืกื•ืจืชื™,
07:20
calculated robotic brain.
162
440593
1229
ืžื•ื— ืžื—ื•ืฉื‘ ื•ืจื•ื‘ื•ื˜ื™.
07:21
It waits for its turn, it analyzes everything, it plans.
163
441846
2640
ื”ื•ื ืžื—ื›ื” ืœืชื•ืจื•, ืžื ืชื— ื”ื›ืœ, ื”ื•ื ืžืชื›ื ืŸ.
07:24
Let's call it the calculated brain.
164
444510
1682
ื‘ื•ืื• ื ืงืจื ืœื• ื”ืžื•ื— ื”ืžื—ื•ืฉื‘.
ื”ื—ืฆื™ ื”ืฉื ื™ ืงื™ื‘ืœื• ืžื•ื— ืฉื”ื•ื ื™ื•ืชืจ ื”ืžื•ื— ืฉืœ ืฉื—ืงืŸ ื”ื‘ืžื” ื”ืœื•ืงื— ืกื™ื›ื•ื ื™ื.
07:26
The other got more the stage actor, risk-taker brain.
165
446216
3247
07:29
Let's call it the adventurous brain.
166
449487
2122
ื‘ื•ืื• ื ืงืจื ืœื• ื”ืžื•ื— ื”ื”ืจืคืชืงื ื™.
07:31
It sometimes acts without knowing everything it has to know.
167
451633
2937
ื”ื•ื ืœืคืขืžื™ื ืคื•ืขืœ ืžื‘ืœื™ ืœื“ืขืช ื›ืœ ืžื” ืฉื™ืฉ ืœื“ืขืช.
07:34
It sometimes makes mistakes and corrects them.
168
454594
2292
ื”ื•ื ืœืคืขืžื™ื ืขื•ืฉื” ื˜ืขื•ื™ื•ืช ื•ืžืชืงืŸ ืื•ืชืŸ.
07:36
And I had them do this very tedious task that took almost 20 minutes,
169
456910
3918
ื•ื ืชืชื™ ืœื”ื ืœื‘ืฆืข ืžืฉื™ืžื” ืžืื•ื“ ืžื™ื™ื’ืขืช
ืฉืœืงื—ื” ื›ืžืขื˜ 20 ื“ืงื•ืช
07:40
and they had to work together,
170
460852
1476
ื•ื”ื ื”ื™ื• ื—ื™ื™ื‘ื™ื ืœืขื‘ื•ื“ ื™ื—ื“,
07:42
somehow simulating, like, a factory job
171
462352
2553
ื‘ืžื™ื“ื” ืžืกื•ื™ื™ืžืช ืžื“ืžื™ื ืžืขื™ื™ืŸ ืขื‘ื•ื“ื” ืฉืœ ืžืคืขืœ
07:44
of repetitively doing the same thing.
172
464929
2087
ืฉืœ ืขืฉื™ื™ืช ืื•ืชื• ื”ื“ื‘ืจ ื‘ืื•ืคืŸ ื—ื•ื–ืจื ื™.
07:47
What I found is that people actually loved the adventurous robot.
173
467445
3331
ื•ืžื” ืฉื’ื™ืœื™ืชื™ ื”ื™ื” ืฉืื ืฉื™ื ื“ื•ื•ืงื ืื”ื‘ื•
ืืช ื”ืจื•ื‘ื•ื˜ ื”ื”ืจืคืชืงื ื™.
07:50
They thought it was more intelligent,
174
470800
1777
ื•ื”ื ื—ืฉื‘ื• ืฉื”ื•ื ื”ื™ื” ื™ื•ืชืจ ืื™ื ื˜ืœื™ื’ื ื˜ื™,
07:52
more committed, a better member of the team,
175
472601
2106
ื™ื•ืชืจ ืžื—ื•ื™ื™ื‘, ื—ื‘ืจ ืฆื•ื•ืช ื™ื•ืชืจ ื˜ื•ื‘,
07:54
contributed to the success of the team more.
176
474731
2065
ืฉืชืจื ืœื”ืฆืœื—ื” ืฉืœ ื”ืฆื•ื•ืช ื™ื•ืชืจ.
07:56
They even called it "he" and "she,"
177
476820
1708
ื”ื ืืคื™ืœื• ืงืจืื• ืœื• ื‘ื›ื™ื ื•ื™ื™ื "ื”ื•ื" ื•"ื”ื™ื",
ื‘ืขื•ื“ ืฉืื ืฉื™ื ืขื ื”ืžื•ื— ื”ืžื—ื•ืฉื‘ ืงืจืื• ืœื• "ื–ื”".
07:58
whereas people with the calculated brain
178
478552
2121
08:00
called it "it," and nobody ever called it "he" or "she."
179
480697
2802
ื•ืืฃ ืื—ื“ ืœื ืงืจื ืœื• "ื”ื•ื" ืื• "ื”ื™ื".
08:03
When they talked about it after the task, with the adventurous brain,
180
483876
3293
ื›ืฉื”ื ื“ื™ื‘ืจื• ืขืœ ื›ืš ืœืื—ืจ ื”ืžืฉื™ืžื”
ืขื ื”ืžื•ื— ื”ื”ืจืคืชืงื ื™,
08:07
they said, "By the end, we were good friends and high-fived mentally."
181
487193
3783
ื”ื ืืžืจื•, "ืœืงืจืืช ื”ืกื•ืฃ ื›ื‘ืจ ื”ื™ื™ื ื• ื—ื‘ืจื™ื ื˜ื•ื‘ื™ื ื•ื ืชื ื• ืื—ื“ ืœืฉื ื™ ื›ื™ืฃ ืžื ื˜ืœื™".
08:11
Whatever that means.
182
491397
1293
ืžื” ืฉื–ื” ืœื ืื•ืžืจ.
08:12
(Laughter)
183
492714
1906
(ืฆื—ื•ืง) ื ืฉืžืข ื›ื•ืื‘.
08:14
Sounds painful.
184
494644
1353
ื‘ืขื•ื“ ื”ืื ืฉื™ื ืขื ื”ืžื•ื— ื”ืžื—ื•ืฉื‘
08:16
Whereas the people with the calculated brain
185
496021
2738
08:18
said it was just like a lazy apprentice.
186
498783
2358
ืืžืจื• ืฉื”ื•ื ื”ื™ื” ื›ืžื• ืฉื•ืœื™ื™ื” ืขืฆืœืŸ.
08:21
It only did what it was supposed to do and nothing more,
187
501165
2745
ื”ื•ื ืขืฉื” ืจืง ืžื” ืฉื”ื•ื ื”ื™ื” ืืžื•ืจ ืœืขืฉื•ืช ื•ืœื ื™ื•ืชืจ.
08:23
which is almost what people expect robots to do,
188
503934
2349
ืฉื–ื” ื›ืžืขื˜ ืžื” ืฉืื ืฉื™ื ืžืฆืคื™ื ืžืจื•ื‘ื•ื˜ื™ื ืœืขืฉื•ืช,
08:26
so I was surprised that people had higher expectations of robots
189
506307
3555
ืื– ื”ื™ื™ืชื™ ืžื•ืคืชืข ืฉืœืื ืฉื™ื ื”ื™ื• ืฆื™ืคื™ื•ืช ื’ื‘ื•ื”ื•ืช ื™ื•ืชืจ ืžื”ืจื•ื‘ื•ื˜ื™ื
ืžืืฉืจ ืžื” ืฉื›ืœ ืื—ื“ ื‘ืชืขืฉื™ื™ืช ื”ืจื•ื‘ื•ื˜ื™ืงื” ื—ืฉื‘ ืฉืจื•ื‘ื•ื˜ื™ื ืฆืจื™ื›ื™ื ืœืขืฉื•ืช.
08:29
than what anybody in robotics thought robots should be doing.
190
509886
3506
ื•ื‘ืžื•ื‘ืŸ ื›ืœืฉื”ื•, ื—ืฉื‘ืชื™, ืื•ืœื™ ื”ื’ื™ืข ื”ื–ืžืŸ--
08:34
And in a way, I thought, maybe it's time --
191
514027
2015
ื‘ื“ื™ื•ืง ื›ืžื• ืฉืžืฉื—ืง ืžืชื•ื“ื™ ืฉื™ื ื” ืืช ื”ืื•ืคืŸ
08:36
just like method acting changed the way people thought
192
516066
2962
ืฉื‘ื• ืื ืฉื™ื ื—ืฉื‘ื• ืขืœ ืžืฉื—ืง ื‘ืžืื” ื”-19,
08:39
about acting in the 19th century,
193
519052
1596
08:40
from going from the very calculated, planned way of behaving,
194
520672
3077
ืžืœืœื›ืช ืžื”ืฆื•ืจื” ื”ืžืื•ื“ ืžื—ื•ืฉื‘ืช,
ื”ืžืชื•ื›ื ื ืช ืฉืœ ื”ื”ืชื ื”ื’ื•ืช,
08:43
to a more intuitive, risk-taking, embodied way of behaving --
195
523773
3498
ืœืฆื•ืจื” ื™ื•ืชืจ ืื™ื ื˜ื•ืื™ื˜ื™ื‘ื™ืช, ืœื•ืงื—ืช ืกื™ื›ื•ื ื™ื, ื ื˜ื•ืขืช ื’ื•ืฃ ืฉืœ ื”ืชื ื”ื’ื•ืช.
08:47
maybe it's time for robots to have the same kind of revolution.
196
527295
3196
ืื•ืœื™ ื–ื” ื”ื–ืžืŸ ืฉืœืจื•ื‘ื•ื˜ื™ื
ืชื”ื™ื” ืืช ืื•ืชื” ืกื•ื’ ืฉืœ ืžื”ืคื™ื›ื”.
08:51
A few years later, I was at my next research job at Georgia Tech in Atlanta,
197
531994
3599
ื›ืžื” ืฉื ื™ื ืžืื•ื—ืจ ื™ื•ืชืจ,
ื”ื™ื™ืชื™ ื‘ืขื‘ื•ื“ืช ื”ืžื—ืงืจ ื”ื‘ืื” ืฉืœื™ ื‘ื’'ื•ืจื’'ื™ื” ื˜ืง ื‘ืื˜ืœื ื˜ื”,
08:55
and I was working in a group dealing with robotic musicians.
198
535617
2824
ื•ืขื‘ื“ืชื™ ื‘ืงื‘ื•ืฆื”
ืฉืขืกืงื” ื‘ืžื•ื–ื™ืงืื™ื ืจื•ื‘ื•ื˜ื™ื™ื.
ื•ืื ื™ ื—ืฉื‘ืชื™, ืžื•ื–ื™ืงื”, ื–ื” ื”ืžืงื•ื ื”ืžื•ืฉืœื
08:58
And I thought, music: that's the perfect place
199
538465
2360
09:00
to look at teamwork, coordination, timing, improvisation --
200
540849
4143
ืœื”ืชื‘ื•ื ืŸ ื‘ืขื‘ื•ื“ืช ืฆื•ื•ืช, ืชื™ืื•ื,
ืชื–ืžื•ืŸ, ืืœืชื•ืจ--
ื•ื‘ื“ื™ื•ืง ื”ืฆืœื—ื ื• ืœื’ืจื•ื ืœืจื•ื‘ื•ื˜ ื”ื–ื” ืœื ื’ืŸ ืžืจื™ืžื‘ื”.
09:05
and we just got this robot playing marimba.
201
545016
2443
09:07
And the marimba, for everybody like me,
202
547483
2256
ืžืจื™ืžื‘ื”, ืขื‘ื•ืจ ื›ืœ ืžื™ ืฉื”ื™ื” ื›ืžื•ื ื™,
09:09
it was this huge, wooden xylophone.
203
549763
2739
ื–ื” ื”ื™ื” ืงืกื™ืœื•ืคื•ืŸ ืขืฆื•ื ืขืฉื•ื™ ืžืขืฅ.
09:12
And when I was looking at this,
204
552526
2063
ื•ื›ืืฉืจ ื”ืชืกืชื›ืœืชื™ ืขืœ ื–ื”,
09:14
I looked at other works in human-robot improvisation --
205
554613
2998
ื”ืกืชื›ืœืชื™ ืขืœ ืขื‘ื•ื“ื•ืช ืื—ืจื•ืช ื‘ืชื—ื•ื ื”ืืœืชื•ืจ ื”ืจื•ื‘ื•ื˜ื™-ืื ื•ืฉื™--
09:17
yes, there are other works in human-robot improvisation --
206
557635
2747
ื›ืŸ, ื™ืฉื ืŸ ืขื•ื“ ืขื‘ื•ื“ื•ืช ื‘ืชื—ื•ื ื”ืืœืชื•ืจ ื”ืจื•ื‘ื•ื˜ื™-ืื ื•ืฉื™--
09:20
and they were also a little bit like a chess game.
207
560406
2349
ื•ื”ืŸ ื’ื ื”ื™ื• ืงืฆืช ื›ืžื• ืžืฉื—ืง ืฉื—ืžื˜.
09:22
The human would play,
208
562779
1151
ื”ืื“ื ื”ื™ื” ืžื ื’ืŸ,
09:23
the robot analyzed what was played,
209
563954
2173
ื”ืจื•ื‘ื•ื˜ ื”ื™ื” ืžื ืชื— ืืช ืžื” ืฉื ื•ื’ืŸ,
ื•ื”ื™ื” ืžืืœืชืจ ืืช ื”ื—ืœืง ืฉืœื•.
09:26
and would improvise their own part.
210
566151
2079
09:28
So, this is what musicians called a call-and-response interaction,
211
568254
3138
ืื–, ื–ื” ืžื” ืฉืžื•ื–ื™ืงืื™ื ืงื•ืจืื™ื ืœื•
ืื™ื ื˜ืจืืงืฆื™ื” ืฉืœ ืงืจื™ืื” ื•ืชื’ื•ื‘ื”.
09:31
and it also fits very well robots and artificial intelligence.
212
571416
3747
ื•ื–ื” ื’ื ืžืฉืชืœื‘ ื”ื™ื˜ื‘, ืจื•ื‘ื•ื˜ื™ื ื•ืื™ื ื˜ืœื™ื’ื ืฆื™ื” ืžืœืื›ื•ืชื™ืช.
09:35
But I thought, if I use the same ideas I used in the theater play
213
575187
3253
ืื‘ืœ ืื ื™ ื—ืฉื‘ืชื™, ืื ืืฉืชืžืฉ ื‘ืื•ืชื ืจืขื™ื•ื ื•ืช ืฉื”ืฉืชืžืฉืชื™ ื‘ื”ื
ื‘ืžื—ื–ื” ื”ืชื™ืื˜ืจื•ืŸ ื•ื‘ืžื—ืงืจื™ ืขื‘ื•ื“ืช ื”ืฆื•ื•ืช,
09:38
and in the teamwork studies,
214
578464
1744
09:40
maybe I can make the robots jam together like a band.
215
580232
3650
ืื•ืœื™ ืื ื™ ื™ื›ื•ืœ ืœื’ืจื•ื ืœืจื•ื‘ื•ื˜ื™ื ืœื’'ืžื’'ื ื‘ื™ื—ื“
ื›ืžื• ืœื”ืงื”.
09:43
Everybody's riffing off each other, nobody is stopping for a moment.
216
583906
4268
ื›ื•ืœื ืฉื•ืื‘ื™ื ื”ืฉืจืื” ืื—ื“ ืžื”ืฉื ื™, ืืฃ ืื—ื“ ืœื ืขื•ืฆืจ ืœืจื’ืข.
09:48
And so I tried to do the same things, this time with music,
217
588198
2874
ื•ื›ืš, ื ื™ืกื™ืชื™ ืœืขืฉื•ืช ืืช ืื•ืชื ื”ื“ื‘ืจื™ื, ื”ืคืขื ืขื ืžื•ื–ื™ืงื”,
ื›ืืฉืจ ื”ืจื•ื‘ื•ื˜ ืœื ื‘ืืžืช ื™ื•ื“ืข
09:51
where the robot doesn't really know what it's about to play,
218
591096
2834
ืžื” ื”ื•ื ืขื•ืžื“ ืœื ื’ืŸ.
09:53
it just sort of moves its body and uses opportunities to play,
219
593954
3017
ื”ื•ื ืคืฉื•ื˜ ืžื–ื™ื– ืืช ื”ื’ื•ืฃ ืฉืœื•
ื•ืžืฉืชืžืฉ ื‘ื”ื–ื“ืžื ื•ื™ื•ืช ืœื ื’ืŸ,
09:56
and does what my jazz teacher when I was 17 taught me.
220
596995
2539
ื•ืขื•ืฉื” ืžื” ืฉื”ืžื•ืจื” ืฉืœื™ ืœื’'ืื– ื›ืฉื”ื™ื™ืชื™ ื‘ืŸ 17 ืœื™ืžื“ื” ืื•ืชื™.
09:59
She said, when you improvise,
221
599558
1397
ื”ื™ื ืืžืจื”, ื›ืฉืืชื” ืžืืœืชืจ,
10:00
sometimes you don't know what you're doing, and you still do it.
222
600979
3016
ืœืคืขืžื™ื ืืชื” ืœื ื™ื•ื“ืข ืžื” ืืชื” ืขื•ืฉื”
ื•ื‘ื›ืœ ื–ืืช ืืชื” ืขื•ืฉื” ืืช ื–ื”.
ื•ื›ืš ื ื™ืกื™ืชื™ ืœื™ืฆื•ืจ ืจื•ื‘ื•ื˜ ืฉืœื ื‘ืืžืช
10:04
So I tried to make a robot that doesn't actually know what it's doing,
223
604019
3326
ื™ื•ื“ืข ืžื” ื”ื•ื ืขื•ืฉื”, ืื‘ืœ ื”ื•ื ื‘ื›ืœ ื–ืืช ืขื•ืฉื” ืืช ื–ื”.
ืื– ื‘ื•ืื• ื ืกืชื›ืœ ืขืœ ื›ืžื” ืฉื ื™ื•ืช ืžื”ื”ื•ืคืขื” ื”ื–ืืช.
10:07
but is still doing it.
224
607369
1221
10:08
So let's look at a few seconds from this performance,
225
608614
2511
ื›ืฉื”ืจื•ื‘ื•ื˜ ืžืงืฉื™ื‘ ืœืžื•ื–ื™ืงืื™ ื”ืื ื•ืฉื™
10:11
where the robot listens to the human musician
226
611149
2127
ื•ืžืืœืชืจ.
10:13
and improvises.
227
613300
1165
10:14
And then, look how the human musician also responds
228
614489
2975
ื•ืื–, ื”ื‘ื™ื˜ื• ืื™ืš ื”ืžื•ื–ื™ืงืื™ ื”ืื ื•ืฉื™ ื’ื
ืžื’ื™ื‘ ืœืžื” ืฉื”ืจื•ื‘ื•ื˜ ืขื•ืฉื”, ื•ืœื•ืงื— ืงืฆืช
10:17
to what the robot is doing
229
617488
1273
10:18
and picking up from its behavior,
230
618785
2644
ืžื”ื”ืชื ื”ื’ื•ืช ืฉืœื•.
10:21
and at some point can even be surprised by what the robot came up with.
231
621453
3956
ื•ื‘ื ืงื•ื“ื” ืžืกื•ื™ื™ืžืช ื™ื›ื•ืœ ืืคื™ืœื• ืœื”ื™ื•ืช ืžื•ืคืชืข ืžืžื” ืฉื”ืจื•ื‘ื•ื˜ ื”ืžืฆื™ื.
10:25
(Music)
232
625433
2194
(ืžื•ื–ื™ืงื”)
11:07
(Music ends)
233
667618
2000
11:11
(Applause)
234
671134
5734
(ืžื—ื™ืื•ืช ื›ืคื™ื™ื)
11:16
Being a musician is not just about making notes,
235
676892
2250
ืœื”ื™ื•ืช ืžื•ื–ื™ืงืื™ ื–ื” ืœื ืจืง ื›ืชื™ื‘ืช ืชื•ื•ื™ื,
11:19
otherwise nobody would ever go see a live show.
236
679166
2348
ืื—ืจืช ืืฃ ืื—ื“ ืืฃ ืคืขื ืœื ื”ื™ื” ื”ื•ืœืš ืœืจืื•ืช ื”ื•ืคืขื” ื—ื™ื”.
11:21
Musicians also communicate with their bodies,
237
681538
2122
ืžื•ื–ื™ืงืื™ื ื’ื ืžืชืงืฉืจื™ื ืขื ื”ื’ื•ืฃ ืฉืœื”ื,
11:23
with other band members, with the audience,
238
683684
2017
ืขื ื—ื‘ืจื™ ืœื”ืงื” ืื—ืจื™ื, ืขื ื”ืงื”ืœ,
11:25
they use their bodies to express the music.
239
685725
2014
ื”ื ืžืฉืชืžืฉื™ื ื‘ื’ื•ืฃ ืฉืœื”ื ืขืœ ืžื ืช ืœื‘ื˜ื ืืช ื”ืžื•ื–ื™ืงื”.
11:27
And I thought, we already have a robot musician on stage,
240
687763
2697
ื•ืื ื™ ื—ืฉื‘ืชื™, ื›ื‘ืจ ื™ืฉ ืœื ื• ืจื•ื‘ื•ื˜ ืžื•ื–ื™ืงืื™ ืขืœ ื”ื‘ืžื”,
ืœืžื” ืœื ืœื”ืคื•ืš ืื•ืชื• ืœืžื•ื–ื™ืงืื™ ื‘ื•ื’ืจ ื•ืฉืœื.
11:30
why not make it be a full-fledged musician?
241
690484
2312
11:32
And I started designing a socially expressive head
242
692820
2674
ื•ื”ืชื—ืœืชื™ ืœืขืฆื‘ ืจืืฉ ื‘ืขืœ ื™ื›ื•ืœืช ื”ื‘ืขื” ื—ื‘ืจืชื™ืช
11:35
for the robot.
243
695518
1343
ืขื‘ื•ืจ ื”ืจื•ื‘ื•ื˜.
11:36
The head doesnโ€™t actually touch the marimba,
244
696885
2070
ื”ืจืืฉ ืœืžืขืฉื” ืœื ื ื•ื’ืข ื‘ืžืจื™ืžื‘ื”,
11:38
it just expresses what the music is like.
245
698979
1959
ื”ื•ื ืคืฉื•ื˜ ืžื‘ื™ืข ืืช ื”ืžื•ื–ื™ืงื”.
11:40
These are some napkin sketches from a bar in Atlanta
246
700962
2496
ืืœื• ืกืงื™ืฆื•ืช ืขืœ ืžืคื™ืช ืžื‘ืจ ื‘ืื˜ืœื ื˜ื”,
11:43
that was dangerously located exactly halfway
247
703482
2543
ืฉื”ื™ื” ืžืžื•ืงื ื‘ืื•ืคืŸ ืžืกื•ื›ืŸ ื‘ื“ื™ื•ืง ื‘ืžื—ืฆื™ืช ื”ื“ืจืš
11:46
between my lab and my home.
248
706049
1811
ื‘ื™ืŸ ื”ืžืขื‘ื“ื” ืฉืœื™ ื•ื”ื‘ื™ืช ืฉืœื™. (ืฆื—ื•ืง)
11:47
So I spent, I would say, on average, three to four hours a day there.
249
707884
3938
ืื– ื‘ื™ืœื™ืชื™, ื”ื™ื™ืชื™ ืื•ืžืจ ื‘ืžืžื•ืฆืข,
ืฉืœื•ืฉ ืขื“ ืืจื‘ืข ืฉืขื•ืช ื‘ื™ื•ื ืฉื.
11:51
I think.
250
711846
1187
ืื ื™ ื—ื•ืฉื‘. (ืฆื—ื•ืง)
11:53
(Laughter)
251
713057
1527
11:54
And I went back to my animation tools and tried to figure out
252
714917
2882
ื•ื›ืืฉืจ ื—ื–ืจืชื™ ืœื›ืœื™ื ืฉืœ ื”ืื ื™ืžืฆื™ื” ืฉืœื™ ื•ื ื™ืกื™ืชื™ ืœื”ื‘ื™ืŸ
11:57
not just what a robotic musician would look like,
253
717823
2370
ืœื ืจืง ืื™ืš ืจื•ื‘ื•ื˜ ืžื•ื–ื™ืงืื™ ืจื•ื‘ื•ื˜ื™ ืฆืจื™ืš ืœื”ื™ืจืื•ืช,
ืืœื ื‘ืขื™ืงืจ ืื™ืš ืžื•ื–ื™ืงืื™ ืจื•ื‘ื•ื˜ื™ ืฆืจื™ืš ืœื”ืชื ื•ืขืข.
12:00
but especially what a robotic musician would move like,
254
720217
2625
12:02
to sort of show that it doesn't like what the other person is playing --
255
722866
3402
ื›ื“ื™ ืกื•ื’ ืฉืœ ืœื”ืจืื•ืช ืฉื”ื•ื ืœื ืื•ื”ื‘ ืืช ืžื” ืฉื”ืื“ื ื”ืื—ืจ ืžื ื’ืŸ--
12:06
and maybe show whatever beat it's feeling at the moment.
256
726292
3825
ื•ืื•ืœื™ ืœื”ืจืื•ืช ืื™ื–ื” ืงืฆื‘ ื”ื•ื ืžืจื’ื™ืฉ
ื‘ืื•ืชื• ื”ืจื’ืข.
12:10
So we ended up actually getting the money to build this robot, which was nice.
257
730141
4503
ืื– ื‘ืกื•ืฃ ื”ืฉื’ื ื• ืืช ื”ื›ืกืฃ ืœื‘ื ื™ื™ืช ื”ืจื•ื‘ื•ื˜ ื”ื–ื”, ืžื” ืฉื”ื™ื” ื ื—ืžื“.
12:14
I'm going to show you now the same kind of performance,
258
734668
2604
ืื ื™ ื”ื•ืœืš ืœื”ืจืื•ืช ืœื›ื ืขื›ืฉื™ื• ืืช ืื•ืชื• ืกื•ื’ ืฉืœ ื”ื•ืคืขื”,
ื”ืคืขื ืขื ืจืืฉ ื‘ืขืœ ื™ื›ื•ืœ ื”ื‘ืขื” ื—ื‘ืจืชื™ืช.
12:17
this time with a socially expressive head.
259
737296
2040
12:19
And notice one thing --
260
739360
1800
ื•ืฉื™ืžื• ืœื‘ ืœื“ื‘ืจ ืื—ื“--
12:21
how the robot is really showing us
261
741184
1671
ื”ืจื•ื‘ื•ื˜ ืžืžืฉ ืžืจืื” ืœื ื•
12:22
the beat it's picking up from the human,
262
742879
1906
ืืช ื”ืงืฆื‘ ืฉื”ื•ื ืงื•ืœื˜ ืžื‘ืŸ ื”ืื“ื.
12:24
while also giving the human a sense that the robot knows what it's doing.
263
744809
3927
ืื ื—ื ื• ื’ื ื ื•ืชื ื™ื ืœืื“ื ืชื—ื•ืฉื” ืฉื”ืจื•ื‘ื•ื˜ ื™ื•ื“ืข ืžื” ื”ื•ื ืขื•ืฉื”.
12:28
And also how it changes the way it moves
264
748760
1920
ื•ื’ื ืืš ื”ื•ื ืžืฉื ื” ืืช ื”ืื•ืคืŸ ืฉื‘ื• ื”ื•ื ืžืชื ื•ืขืข
12:30
as soon as it starts its own solo.
265
750704
1952
ื‘ืจื’ืข ืฉื”ื•ื ืžืชื—ื™ืœ ืกื•ืœื• ืžืฉืœื•
12:32
(Music)
266
752680
3896
(ืžื•ื–ื™ืงื”)
12:36
Now it's looking at me, showing that it's listening.
267
756600
3020
ืขื›ืฉื™ื• ื”ื•ื ืžืกืชื›ืœ ืขืœื™ ืœื•ื•ื“ื ืฉืื ื™ ืžืงืฉื™ื‘.
12:39
(Music)
268
759644
2723
(ืžื•ื–ื™ืงื”)
ื•ืขื›ืฉื™ื• ื”ื‘ื™ื˜ื• ื‘ืืงื•ืจื“ ื”ืื—ืจื•ืŸ ืฉืœ ื”ื™ืฆื™ืจื” ืฉื•ื‘,
13:01
Now look at the final chord of the piece again.
269
781181
2921
13:04
And this time the robot communicates with its body
270
784126
2917
ื•ื”ืคืขื ื”ืจื•ื‘ื•ื˜ ืžืชืงืฉืจ ืขื ื”ื’ื•ืฃ ืฉืœื•
13:07
when it's busy doing its own thing,
271
787067
2113
ื›ืฉื”ื•ื ืขืกื•ืง ื‘ืœืขืฉื•ืช ืืช ื”ืงื˜ืข ืฉืœื•.
13:09
and when it's ready to coordinate the final chord with me.
272
789204
4851
ื•ื›ืฉื”ื•ื ืžื•ื›ืŸ
ืœืชืื ืืช ื”ืืงื•ืจื“ ื”ืื—ืจื•ืŸ ืื™ืชื™.
13:14
(Music)
273
794079
2737
(ืžื•ื–ื™ืงื”)
13:21
(Music ending)
274
801783
2000
13:26
(Final chord)
275
806103
1001
13:27
(Applause)
276
807128
5754
(ืžื—ื™ืื•ืช ื›ืคื™ื™ื)
13:32
Thanks.
277
812906
1155
ืชื•ื“ื”. ืื ื™ ืžืงื•ื•ื” ืฉืืชื ืจื•ืื™ื ืขื“ ื›ืžื”--
13:34
I hope you see
278
814085
1581
13:35
how much this part of the body that doesn't touch the instrument
279
815690
4396
ืขื“ ื›ืžื” ื”ื—ืœืง ื”ื–ื” ืฉืœ ื”ื’ื•ืฃ ืฉืœื ื ื•ื’ืข ื‘ื›ืœื™
ืœืžืขืฉื” ืขื•ื–ืจ ืขื ื”ื”ื•ืคืขื” ื”ืžื•ื–ื™ืงืœื™ืช.
13:40
actually helps with the musical performance.
280
820110
2592
13:43
And at some point -- we are in Atlanta,
281
823160
2009
ื•ื‘ื ืงื•ื“ื” ืžืกื•ื™ื™ืžืช, ืื ื—ื ื• ื‘ืื˜ืœื ื˜ื”, ืื– ื‘ืจื•ืจ ืฉืื™ื–ื” ืจืืคืจ
13:45
so obviously some rapper will come into our lab at some point --
282
825193
3250
ื™ื‘ื•ื ืœืžืขื‘ื“ื” ืฉืœื ื• ื‘ืฉืœื‘ ืžืกื•ื™ื™ื.
13:48
and we had this rapper come in and do a little jam with the robot.
283
828467
4793
ื•ืื™ื–ื” ืจืืคืจ ื‘ื
ื•ืขืฉื” ื’'ืื ืงื˜ืŸ ืขื ื”ืจื•ื‘ื•ื˜.
13:53
Here you can see the robot basically responding to the beat.
284
833284
4048
ื•ื›ืืŸ ืืคืฉืจ ืœืจืื•ืช ืืช ื”ืจื•ื‘ื•ื˜
ืœืžืขืฉื” ืžื’ื™ื‘ ืœืงืฆื‘ ื•--
13:57
Notice two things:
285
837356
1151
ืฉื™ืžื• ืœื‘ ืœืฉื ื™ ื“ื‘ืจื™ื, ืื—ื“, ื›ืžื” ืงืฉื” ืœื”ืชื ื’ื“
13:58
one, how irresistible it is to join the robot while it's moving its head.
286
838531
3609
ืœื”ืฆื˜ืจืฃ ืœืจื•ื‘ื•ื˜ ื›ืฉื”ื•ื ืžื ื™ืข ืืช ืจืืฉื•.
14:02
You kind of want to move your own head when it does it.
287
842164
2603
ื•ืืชื” ืกื•ื’ ืฉืœ ืจื•ืฆื” ืœื”ื–ื™ื– ืืช ื”ืจืืฉ ืฉืœืš ื›ืฉื”ื•ื ืขื•ืฉื” ื–ืืช.
14:04
And second, even though the rapper is really focused on his iPhone,
288
844791
3586
ื•ื“ื‘ืจ ืฉื ื™, ืœืžืจื•ืช ืฉื”ืจืืคืจ ื‘ืขืฆื ืžืจื•ื›ื– ื‘ืื™ื™ืคื•ืŸ ืฉืœื•,
14:08
as soon as the robot turns to him, he turns back.
289
848401
3110
ื‘ืจื’ืข ืฉื”ืจื•ื‘ื•ื˜ ืžืกืชื•ื‘ื‘ ืืœื™ื•, ื”ื•ื ืžืกืชื•ื‘ื‘ ื‘ื—ื–ืจื”.
14:11
So even though it's just in the periphery of his vision,
290
851535
2636
ืื– ืืคื™ืœื• ืฉื–ื” ืจืง ื‘ืคืจืคืจื™ื” ืฉืœ ื”ืจืื™ื” ืฉืœื•--
ื–ื” ืจืง ื‘ืคื™ื ืช ื”ืขื™ืŸ ืฉืœื•-- ื–ื” ืžืื•ื“ ื—ื–ืง.
14:14
in the corner of his eye, it's very powerful.
291
854195
2120
ื•ื”ืกื™ื‘ื” ื”ื™ื ืฉืื ื—ื ื• ืœื ื™ื›ื•ืœื™ื ืœื”ืชืขืœื
14:16
And the reason is that we can't ignore
292
856339
1834
ืžื“ื‘ืจื™ื ืคื™ืกื™ื™ื ืฉื–ื–ื™ื ื‘ืกื‘ื™ื‘ื” ืฉืœื ื•.
14:18
physical things moving in our environment.
293
858197
2024
ืื ื—ื ื• ืžื—ื•ื•ื˜ื™ื ื›ื›ื”.
14:20
We are wired for that.
294
860245
1170
14:21
So if you have a problem --
295
861439
1644
ืื–, ืื ื™ืฉ ืœื›ื ื‘ืขื™ื” ืขื ื ืืžืจ ื–ื” ืฉื”ืฉื•ืชืคื™ื ืฉืœื›ื
14:23
maybe your partner is looking at their iPhone or smartphone too much --
296
863107
4675
ืžืกืชื›ืœื™ื ืขืœ ื”ืื™ื™ืคื•ื ื™ื ืฉืœื”ื ื™ื•ืชืจ ืžื™ื“ื™, ืื• ื”ืกืžืืจื˜ืคื•ื ื™ื ืฉืœื”ื ื™ื•ืชืจ ืžื“ื™,
14:27
you might want to have a robot there to get their attention.
297
867806
2834
ืื•ืœื™ ืชืจืฆื• ืฉื™ื”ื™ื” ืฉื ืจื•ื‘ื•ื˜
ืฉื™ืฉื™ื’ ืืช ืชืฉื•ืžืช ื”ืœื‘ ืฉืœื”ื. (ืฆื—ื•ืง)
14:30
(Laughter)
298
870664
1001
(ืžื•ื–ื™ืงื”)
14:31
(Music)
299
871689
2069
14:46
(Music ends)
300
886113
2000
14:50
(Applause)
301
890138
6948
(ืžื—ื™ืื•ืช ื›ืคื™ื™ื)
ืจืง ื›ื“ื™ ืœื”ืฆื™ื’ ืืช ื”ืจื•ื‘ื•ื˜ ื”ืื—ืจื•ืŸ
14:57
Just to introduce the last robot that we've worked on,
302
897633
3834
ืฉืขื‘ื“ื ื• ืขืœื™ื•,
15:01
it came out of something surprising that we found:
303
901491
2350
ืฉื™ืฆื ืžืžืฉื”ื• ื“ื™ ืžืคืชื™ืข ืฉื’ื™ืœื™ื ื•:
15:03
Some point people didn't care about the robot being intelligent,
304
903865
3158
ื‘ืฉืœื‘ ืžืกื•ื™ื™ื ืœืื ืฉื™ื ืœื ื”ื™ื” ืื›ืคืช ื™ื•ืชืจ ืžื–ื” ืฉื”ืจื•ื‘ื•ื˜ ื›ืœ ื›ืš ืื™ื ื˜ืœื™ื’ื ื˜ื™,
ื•ื™ื›ื•ืœ ืœืืœืชืจ ื•ืœื”ืงืฉื™ื‘,
15:07
able to improvise and listen,
305
907047
1402
15:08
and do all these embodied intelligence things that I spent years developing.
306
908473
4406
ื•ืœืขืฉื•ืช ืืช ื›ืœ ื”ื“ื‘ืจื™ื ืฉืœ ื”ืื™ื ื˜ืœื™ื’ื ืฆื™ื” ื ื˜ื•ืขืช ื”ื’ื•ืฃ ืฉื‘ื™ืœื™ืชื™ ืฉื ื™ื ื‘ืคื™ืชื•ื—ื.
15:12
They really liked that the robot was enjoying the music.
307
912903
2681
ื”ื ืžืžืฉ ืื”ื‘ื• ืืช ื–ื” ืฉื”ืจื•ื‘ื•ื˜ ื ื”ื ื” ืžื”ืžื•ื–ื™ืงื”. (ืฆื—ื•ืง)
15:15
(Laughter)
308
915608
1001
15:16
And they didn't say the robot was moving to the music,
309
916633
2545
ื•ื”ื ืœื ืืžืจื• ืฉื”ืจื•ื‘ื•ื˜ ื ืข ืœืคื™ ื”ืžื•ื–ื™ืงื”,
ื”ื ืืžืจื• ืฉื”ืจื•ื‘ื•ื˜ ื ื”ื ื” ืžื”ืžื•ื–ื™ืงื”.
15:19
they said "enjoying" the music.
310
919202
1488
15:20
And we thought, why don't we take this idea,
311
920714
2126
ื•ืื ื—ื ื• ื—ืฉื‘ื ื•, ืœืžื” ืฉืœื ื ื™ืงื— ืืช ื”ืจืขื™ื•ืŸ ื”ื–ื”,
15:22
and I designed a new piece of furniture.
312
922864
2741
ื•ืขื™ืฆื‘ืชื™ ืคื™ืกืช ืจื™ื”ื•ื˜ ื—ื“ืฉื”.
15:25
This time it wasn't a desk lamp, it was a speaker dock,
313
925629
2572
ื”ืคืขื ื–ืืช ืœื ื”ื™ืชื” ืžื ื•ืจืช ืฉื•ืœื—ืŸ; ื–ืืช ื”ื™ื™ืชื” ืชื—ื ืช ืขื’ื™ื ื” ืขื ืจืžืงื•ืœ.
ื–ื” ื”ื™ื” ืื—ื“ ืžื”ื“ื‘ืจื™ื ื”ืืœื” ืฉืืชื” ืžื—ื‘ืจ ืืช ื”ืกืžืืจื˜ืคื•ืŸ ืฉืœืš ืืœื™ื•.
15:28
one of those things you plug your smartphone in.
314
928225
2640
15:30
And I thought,
315
930889
1150
ื•ื—ืฉื‘ืชื™, ืžื” ื™ืงืจื”
15:32
what would happen if your speaker dock didn't just play the music for you,
316
932063
3500
ืื ื”ืจืžืงื•ืœ ืฉืœืš ืœื ืจืง ื™ื ื’ืŸ ืืช ื”ืžื•ื–ื™ืงื” ืขื‘ื•ืจืš,
15:35
but would actually enjoy it, too?
317
935587
2046
ืืœื ื’ื ืžืžืฉ ื™ื”ื ื” ืžืžื ื”. (ืฆื—ื•ืง)
15:37
And so again, here are some animation tests from an early stage.
318
937657
4161
ื•ื›ืš ืฉื•ื‘, ื”ื ื” ื›ืžื” ืžื‘ื—ื ื™ ืื ื™ืžืฆื™ื”
ืžืฉืœื‘ ืžื•ืงื“ื. (ืฆื—ื•ืง)
15:41
(Laughter)
319
941842
1384
15:43
And this is what the final product looked like.
320
943835
2871
ื•ื›ืš ื ืจืื” ื”ืžื•ืฆืจ ื”ืกื•ืคื™.
15:58
(Music)
321
958247
2181
("ืชืคื™ืœื• ืืช ื–ื” ื›ืื™ืœื• ืฉื–ื” ื—ื")
16:18
(Music ends)
322
978848
2000
16:21
So, a lot of bobbing heads.
323
981461
2620
ืื–, ื”ืจื‘ื” ืจืืฉื™ื ืžืชื ื•ืขืขื™ื ืžืขืœื” ื•ืžื˜ื”.
16:24
(Applause)
324
984105
3611
(ืžื—ื™ืื•ืช ื›ืคื™ื™ื)
16:27
A lot of bobbing heads in the audience,
325
987740
1890
ื”ืจื‘ื” ืจืืฉื™ื ืžืชื ื•ืขืขื™ื ืžืขืœื” ื•ืžื˜ื” ื‘ืงื”ืœ,
16:29
so we can still see robots influence people.
326
989654
2708
ืื– ืขื“ื™ื™ืŸ ืืคืฉืจ ืœืจืื•ืช ืฉืจื•ื‘ื•ื˜ื™ื ืžืฉืคื™ืขื™ื ืขืœ ืื ืฉื™ื.
16:32
And it's not just fun and games.
327
992386
2785
ื•ื–ื” ืœื ืจืง ื›ื™ืฃ ื•ืžืฉื—ืงื™ื.
16:35
I think one of the reasons I care so much
328
995195
2201
ืื ื™ ื—ื•ืฉื‘ ืฉืื—ืช ืžื”ืกื™ื‘ื•ืช ืฉืื›ืคืช ืœื™ ื›ืœ ื›ืš
16:37
about robots that use their body to communicate
329
997420
2233
ืœื’ื‘ื™ ืจื•ื‘ื•ื˜ื™ื ืฉืžืฉืชืžืฉื™ื ื‘ื’ื•ืคื ืขืœ ืžื ืช ืœืชืงืฉืจ
16:39
and use their body to move is --
330
999677
1706
ื•ืžืฉืชืžืฉื™ื ื‘ื’ื•ืคื ืขืœ ืžื ืช ืœื–ื•ื–--
16:41
I'm going to let you in on a little secret we roboticists are hiding --
331
1001407
3390
ื•ืื ื™ ื”ื•ืœืš ืœื’ืœื•ืช ืœื›ื ืกื•ื“ ืงื˜ืŸ ืฉืื ื—ื ื• ื”ืจื•ื‘ื•ื˜ื™ืงืื™ื ืžืกืชื™ืจื™ื--
16:44
is that every one of you is going to be living with a robot
332
1004821
2793
ื”ื™ื ืฉื›ืœ ืื—ื“ ืžื›ื ื”ื•ืœืš ืœื—ื™ื•ืช ืขื ืจื•ื‘ื•ื˜
16:47
at some point in your life.
333
1007638
1592
ื‘ืฉืœื‘ ืžืกื•ื™ื™ื ื‘ื—ื™ื™ื›ื.
16:49
Somewhere in your future, there will be a robot in your life.
334
1009254
2933
ืื™ืคืฉื”ื• ื‘ืขืชื™ื“ ืฉืœื›ื ื”ื•ืœืš ืœื”ื™ื•ืช ืจื•ื‘ื•ื˜ ื‘ื—ื™ื™ื›ื.
ื•ืื ืœื ืฉืœื›ื, ืื– ืฉืœ ื”ื™ืœื“ื™ื ืฉืœื›ื.
16:52
If not in yours, your children's lives.
335
1012211
1868
ื•ืื ื™ ืจื•ืฆื” ืฉื”ืจื•ื‘ื•ื˜ื™ื ื”ืืœื” ื™ื”ื™ื•--
16:54
And I want these robots to be more fluent, more engaging, more graceful
336
1014103
4990
ื™ื”ื™ื• ื™ื•ืชืจ ื–ื•ืจืžื™ื, ื™ื•ืชืจ ื›ื•ื‘ืฉื™ื, ื™ื•ืชืจ ื—ื™ื ื ื™ื™ื
ืžืื™ืš ืฉื”ื ื›ื™ื•ื.
16:59
than currently they seem to be.
337
1019117
1729
17:00
And for that I think maybe robots need to be less like chess players
338
1020870
3287
ื•ืœืฉื ื›ืš ืื ื™ ื—ื•ืฉื‘ ืฉืจื•ื‘ื•ื˜ื™ื ืื•ืœื™
ืฆืจื™ื›ื™ื ืœื”ื™ื•ืช ืคื—ื•ืช ื›ืžื• ืฉื—ืงื ื™ ืฉื—ืžื˜
17:04
and more like stage actors and more like musicians.
339
1024181
2641
ื•ื™ื•ืชืจ ื›ืžื• ืฉื—ืงื ื™ ื‘ืžื” ื•ืžื•ื–ื™ืงืื™ื.
17:06
Maybe they should be able to take chances and improvise.
340
1026846
2813
ืื•ืœื™ ื”ื ืฆืจื™ื›ื™ื ืœื”ื™ื•ืช ืžืกื•ื’ืœื™ื ืœืงื—ืช ืฆ'ืื ืกื™ื ื•ืœืืœืชืจ.
17:09
Maybe they should be able to anticipate what you're about to do.
341
1029683
3024
ื•ืื•ืœื™ ื”ื ืฆืจื™ื›ื™ื ืœื”ื™ื•ืช ืžืกื•ื’ืœื™ื ืœืฆืคื•ืช ืืช ืžื” ืฉืืชื ืขื•ืžื“ื™ื ืœืขืฉื•ืช.
17:12
Maybe they even need to be able to make mistakes and correct them,
342
1032731
3348
ื•ืื•ืœื™ ื”ื ืฆืจื™ื›ื™ื ืœื”ื™ื•ืช ืžืกื•ื’ืœื™ื ืœืขืฉื•ืช ื˜ืขื•ื™ื•ืช
ื•ืœืชืงืŸ ืื•ืชืŸ,
ื›ื™ื•ื•ืŸ ืฉื‘ืกื•ืคื• ืฉืœ ื“ื‘ืจ ืื ื—ื ื• ืื ื•ืฉื™ื™ื.
17:16
because in the end, we are human.
343
1036103
1888
17:18
And maybe as humans, robots that are a little less than perfect
344
1038015
3554
ื•ืื•ืœื™ ื‘ืชื•ืจ ื‘ื ื™ ืื ื•ืฉ, ืจื•ื‘ื•ื˜ื™ื ืฉื”ื ืงืฆืช ืคื—ื•ืช ืžืžื•ืฉืœืžื™ื
17:21
are just perfect for us.
345
1041593
1731
ื”ื ื‘ื“ื™ื•ืง ืžื•ืฉืœืžื™ื ืขื‘ื•ืจื ื•.
17:23
Thank you.
346
1043348
1159
ืชื•ื“ื”.
17:24
(Applause)
347
1044531
6836
(ืžื—ื™ืื•ืช ื›ืคื™ื™ื)

Original video on YouTube.com
ืขืœ ืืชืจ ื–ื”

ืืชืจ ื–ื” ื™ืฆื™ื’ ื‘ืคื ื™ื›ื ืกืจื˜ื•ื ื™ YouTube ื”ืžื•ืขื™ืœื™ื ืœืœื™ืžื•ื“ ืื ื’ืœื™ืช. ืชื•ื›ืœื• ืœืจืื•ืช ืฉื™ืขื•ืจื™ ืื ื’ืœื™ืช ื”ืžื•ืขื‘ืจื™ื ืขืœ ื™ื“ื™ ืžื•ืจื™ื ืžื”ืฉื•ืจื” ื”ืจืืฉื•ื ื” ืžืจื—ื‘ื™ ื”ืขื•ืœื. ืœื—ืฅ ืคืขืžื™ื™ื ืขืœ ื”ื›ืชื•ื‘ื™ื•ืช ื‘ืื ื’ืœื™ืช ื”ืžื•ืฆื’ื•ืช ื‘ื›ืœ ื“ืฃ ื•ื™ื“ืื• ื›ื“ื™ ืœื”ืคืขื™ืœ ืืช ื”ืกืจื˜ื•ืŸ ืžืฉื. ื”ื›ืชื•ื‘ื™ื•ืช ื’ื•ืœืœื•ืช ื‘ืกื ื›ืจื•ืŸ ืขื ื”ืคืขืœืช ื”ื•ื•ื™ื“ืื•. ืื ื™ืฉ ืœืš ื”ืขืจื•ืช ืื• ื‘ืงืฉื•ืช, ืื ื ืฆื•ืจ ืื™ืชื ื• ืงืฉืจ ื‘ืืžืฆืขื•ืช ื˜ื•ืคืก ื™ืฆื™ืจืช ืงืฉืจ ื–ื”.

https://forms.gle/WvT1wiN1qDtmnspy7