Robots with "soul" | Guy Hoffman | TEDxJaffa
-
0:18 - 0:23My job is to design, build and study
robots that communicate with people. -
0:23 - 0:26But this story doesn't start with robotics
at all, it starts with animation. -
0:26 - 0:29When I first saw Pixar's "Luxo Jr.,"
-
0:29 - 0:32I was amazed by how much
emotion they could put -
0:32 - 0:35into something as trivial as a desk lamp.
-
0:35 - 0:38I mean, look at them --
at the end of this movie, -
0:38 - 0:41you actually feel something
for two pieces of furniture. -
0:41 - 0:43And I said, I have
to learn how to do this. -
0:43 - 0:45So I made a really bad career decision.
-
0:45 - 0:46(Laughter)
-
0:46 - 0:48And that's what my mom
was like when I did it. -
0:48 - 0:50(Laughter)
-
0:50 - 0:54I left a very cozy tech job in Israel
at a nice software company -
0:54 - 0:56and I moved to New York
to study animation. -
0:56 - 0:57And there I lived
-
0:57 - 1:00in a collapsing apartment building
in Harlem with roommates. -
1:00 - 1:02I'm not using this phrase
metaphorically -- -
1:02 - 1:05the ceiling actually collapsed
one day in our living room. -
1:05 - 1:08Whenever they did news stories
about building violations in New York, -
1:08 - 1:11they would put the report
in front of our building, -
1:11 - 1:13as kind of, like, a backdrop
to show how bad things are. -
1:13 - 1:15Anyway, during the day, I went to school
-
1:15 - 1:19and at night I would sit and draw
frame by frame of pencil animation. -
1:19 - 1:21And I learned two surprising lessons.
-
1:21 - 1:25One of them was that
when you want to arouse emotions, -
1:25 - 1:28it doesn't matter so much
how something looks; -
1:28 - 1:31it's all in the motion, in the timing
of how the thing moves. -
1:31 - 1:34And the second was something
one of our teachers told us. -
1:34 - 1:37He actually did the weasel in "Ice Age."
-
1:37 - 1:41And he said, "As an animator,
you're not a director -- you're an actor." -
1:41 - 1:44So, if you want to find
the right motion for a character, -
1:44 - 1:47don't think about it --
go use your body to find it. -
1:47 - 1:50Stand in front of a mirror,
act it out in front of a camera -- -
1:50 - 1:53whatever you need -- and then
put it back in your character. -
1:53 - 1:55A year later I found myself at MIT
in the Robotic Life Group. -
1:55 - 1:58It was one of the first groups
researching the relationships -
1:58 - 2:00between humans and robots.
-
2:00 - 2:03And I still had this dream
to make an actual, physical Luxo Jr. lamp. -
2:03 - 2:06But I found that robots didn't move
at all in this engaging way -
2:06 - 2:08that I was used to
from my animation studies. -
2:08 - 2:11Instead, they were all --
how should I put it -- -
2:11 - 2:12they were all kind of robotic.
-
2:12 - 2:14(Laughter)
-
2:14 - 2:17And I thought, what if I took
whatever I learned in animation school, -
2:17 - 2:20and used that to design
my robotic desk lamp. -
2:20 - 2:22So I went and designed frame by frame
-
2:22 - 2:26to try to make this robot as graceful
and engaging as possible. -
2:27 - 2:30And here when you see the robot
interacting with me on a desktop -- -
2:30 - 2:32and I'm actually redesigning the robot,
-
2:32 - 2:34so, unbeknownst to itself,
-
2:34 - 2:37it's kind of digging its own
grave by helping me. -
2:37 - 2:39(Laughter)
-
2:39 - 2:42I wanted it to be less of a mechanical
structure giving me light, -
2:42 - 2:45and more of a helpful,
kind of quiet apprentice -
2:45 - 2:48that's always there when you need it
and doesn't really interfere. -
2:48 - 2:52And when, for example, I'm looking
for a battery that I can't find, -
2:52 - 2:55in a subtle way, it'll show me
where the battery is. -
2:55 - 2:58And I want you to notice
how the same mechanical structure -
2:58 - 3:00can at one point,
just by the way it moves, -
3:00 - 3:01seem gentle and caring,
-
3:01 - 3:04and in the other case seem
violent and confrontational -
3:04 - 3:07And it's the same structure,
just the motion is different. -
3:07 - 3:10Actor: "You want to know something?
-
3:10 - 3:12You wanna know something?!
-
3:12 - 3:13He was already dead!
-
3:14 - 3:18Just laying there, eyes glazed over."
-
3:18 - 3:19(Laughter)
-
3:19 - 3:22But, moving in a graceful way
is just one building block -
3:22 - 3:25of this whole structure
called human-robot interaction. -
3:25 - 3:28I was, at the time, doing my PhD,
I was working on human-robot teamwork, -
3:28 - 3:30teams of humans and robots
working together. -
3:30 - 3:32I was studying the engineering,
-
3:32 - 3:34the psychology,
the philosophy of teamwork, -
3:34 - 3:36and at the same time,
-
3:36 - 3:38I found myself in my own kind
of teamwork situation, -
3:38 - 3:40with a good friend of mine,
who's actually here. -
3:40 - 3:43And in that situation,
we can easily imagine robots -
3:43 - 3:45in the near future being there with us.
-
3:45 - 3:46It was after a Passover Seder.
-
3:46 - 3:48We were folding up
a lot of folding chairs, -
3:48 - 3:51and I was amazed at how quickly
we found our own rhythm. -
3:51 - 3:54Everybody did their own part,
we didn't have to divide our tasks. -
3:54 - 3:57We didn't have to communicate
verbally about this -- -
3:57 - 3:58it all just happened.
-
3:58 - 4:01And I thought, humans and robots
don't look at all like this. -
4:01 - 4:04When humans and robots interact,
it's much more like a chess game: -
4:04 - 4:07the human does a thing, the robot
analyzes whatever the human did, -
4:07 - 4:10the robot decides what to do next,
plans it and does it. -
4:10 - 4:12Then the human waits,
until it's their turn again. -
4:12 - 4:15So it's much more like a chess game,
and that makes sense, -
4:15 - 4:18because chess is great for mathematicians
and computer scientists. -
4:18 - 4:22It's all about information, analysis,
decision-making and planning. -
4:22 - 4:24But I wanted my robot
to be less of a chess player, -
4:24 - 4:26and more like a doer
-
4:26 - 4:28that just clicks and works together.
-
4:28 - 4:31So I made my second
horrible career choice: -
4:31 - 4:34- actually, I think maybe
I should title this talk -
4:35 - 4:38"A lot of ways to give your Jewish mother
a heart attack." -
4:39 - 4:40(Laughter)
-
4:41 - 4:43I decided to study acting for a semester.
-
4:43 - 4:46I took off from the PhD,
I went to acting classes. -
4:46 - 4:49I actually participated in a play --
-
4:49 - 4:51I hope there’s no video
of that around still. -
4:51 - 4:52(Laughter)
-
4:52 - 4:55And I got every book
I could find about acting, -
4:55 - 4:58including one from the 19th century
that I got from the library. -
4:58 - 5:01And I was really amazed, because my name
was the second name on the list -- -
5:01 - 5:03the previous name was in 1889.
-
5:03 - 5:04(Laughter)
-
5:05 - 5:07And this book was
kind of waiting for 100 years -
5:07 - 5:08to be rediscovered for robotics.
-
5:09 - 5:10And this book shows actors
-
5:10 - 5:12how to move every muscle in the body
-
5:12 - 5:15to match every kind of emotion
that they want to express. -
5:15 - 5:18But the real revelation was
when I learned about method acting. -
5:18 - 5:20It became very popular
in the 20th century. -
5:20 - 5:22And method acting said
-
5:22 - 5:24you don't have to plan
every muscle in your body; -
5:24 - 5:27instead, you have to use your body
to find the right movement. -
5:27 - 5:30You have to use your sense memory
to reconstruct the emotions -
5:30 - 5:33and kind of think with your body
to find the right expression -- -
5:33 - 5:35improvise, play off your scene partner.
-
5:35 - 5:36And this came at the same time
-
5:36 - 5:39as I was reading about this trend
in cognitive psychology, -
5:39 - 5:43called embodied cognition,
which also talks about the same ideas. -
5:43 - 5:44We use our bodies to think;
-
5:44 - 5:47we don't just think with our brains
and use our bodies to move, -
5:47 - 5:49but our bodies feed back into our brain
-
5:49 - 5:51to generate the way that we behave.
-
5:51 - 5:53And it was like a lightning bolt.
-
5:53 - 5:54I went back to my office,
-
5:54 - 5:57I wrote this paper,
which I never really published, -
5:57 - 5:59called "Acting Lessons
for Artificial Intelligence." -
5:59 - 6:01And I even took another month
-
6:01 - 6:03to do what was then the first theater play
-
6:03 - 6:05with a human and a robot acting together.
-
6:05 - 6:07That's what you saw
before with the actors. -
6:09 - 6:10And I thought:
-
6:10 - 6:13How can we make an artificial
intelligence model -- -
6:13 - 6:14a computational model --
-
6:14 - 6:17that will model some of these
ideas of improvisation, -
6:17 - 6:19of taking risks, of taking chances,
-
6:19 - 6:20even of making mistakes?
-
6:20 - 6:23Maybe it can make for better
robotic teammates. -
6:23 - 6:26So I worked for quite
a long time on these models -
6:26 - 6:28and I implemented them
on a number of robots. -
6:28 - 6:30Here you can see a very early example
-
6:30 - 6:34with the robots trying to use
this embodied artificial intelligence -
6:34 - 6:36to try to match my movements
as closely as possible. -
6:36 - 6:38It's sort of like a game.
-
6:38 - 6:41And it's a little bit like
what you might see actors do -
6:41 - 6:45when they try to mirror each other
to find the right synchrony between them. -
6:45 - 6:46And then, I did another experiment,
-
6:47 - 6:51and I got people off the street
to use the robotic desk lamp, -
6:51 - 6:54and try out this idea
of embodied artificial intelligence. -
6:54 - 6:56I actually used two kinds
of brains for the same robot. -
6:57 - 6:58The robot is the same lamp that you saw,
-
6:59 - 7:00and I put two brains in it.
-
7:00 - 7:02For one half of the people,
-
7:02 - 7:04I put in a brain
that's kind of the traditional, -
7:04 - 7:05calculated robotic brain.
-
7:05 - 7:08It waits for its turn,
it analyzes everything, it plans. -
7:08 - 7:09Let's call it the calculated brain.
-
7:10 - 7:13The other got more the stage
actor, risk-taker brain. -
7:13 - 7:15Let's call it the adventurous brain.
-
7:15 - 7:18It sometimes acts without knowing
everything it has to know. -
7:18 - 7:20It sometimes makes mistakes
and corrects them. -
7:20 - 7:24And I had them do this very tedious task
that took almost 20 minutes, -
7:24 - 7:26and they had to work together,
-
7:26 - 7:28somehow simulating, like, a factory job
-
7:28 - 7:30of repetitively doing the same thing.
-
7:30 - 7:34What I found is that people
actually loved the adventurous robot. -
7:34 - 7:35They thought it was more intelligent,
-
7:36 - 7:38more committed,
a better member of the team, -
7:38 - 7:40contributed to the success
of the team more. -
7:40 - 7:41They even called it "he" and "she,"
-
7:41 - 7:44whereas people with the calculated brain
-
7:44 - 7:46called it "it," and nobody
ever called it "he" or "she." -
7:47 - 7:50When they talked about it after the task,
with the adventurous brain, -
7:50 - 7:54they said, "By the end, we were good
friends and high-fived mentally." -
7:54 - 7:55Whatever that means.
-
7:55 - 7:57(Laughter)
-
7:57 - 7:58Sounds painful.
-
7:58 - 8:01Whereas the people
with the calculated brain -
8:01 - 8:04said it was just like a lazy apprentice.
-
8:04 - 8:06It only did what it was supposed
to do and nothing more, -
8:06 - 8:09which is almost what people
expect robots to do, -
8:09 - 8:12so I was surprised that people
had higher expectations of robots -
8:12 - 8:16than what anybody in robotics
thought robots should be doing. -
8:17 - 8:19And in a way, I thought,
maybe it's time -- -
8:19 - 8:22just like method acting
changed the way people thought -
8:22 - 8:24about acting in the 19th century,
-
8:24 - 8:27from going from the very calculated,
planned way of behaving, -
8:27 - 8:30to a more intuitive, risk-taking,
embodied way of behaving -- -
8:30 - 8:34maybe it's time for robots
to have the same kind of revolution. -
8:35 - 8:39A few years later, I was at my next
research job at Georgia Tech in Atlanta, -
8:39 - 8:42and I was working in a group
dealing with robotic musicians. -
8:42 - 8:44And I thought, music:
that's the perfect place -
8:44 - 8:48to look at teamwork, coordination,
timing, improvisation -- -
8:49 - 8:51and we just got this robot
playing marimba. -
8:51 - 8:53And the marimba, for everybody like me,
-
8:53 - 8:56it was this huge, wooden xylophone.
-
8:56 - 8:58And when I was looking at this,
-
8:58 - 9:01I looked at other works
in human-robot improvisation -- -
9:01 - 9:04yes, there are other works
in human-robot improvisation -- -
9:04 - 9:06and they were also a little bit
like a chess game. -
9:06 - 9:07The human would play,
-
9:07 - 9:10the robot analyzed what was played,
-
9:10 - 9:12and would improvise their own part.
-
9:12 - 9:15So, this is what musicians called
a call-and-response interaction, -
9:15 - 9:18and it also fits very well
robots and artificial intelligence. -
9:18 - 9:22But I thought, if I use the same ideas
I used in the theater play -
9:22 - 9:23and in the teamwork studies,
-
9:23 - 9:27maybe I can make the robots
jam together like a band. -
9:27 - 9:31Everybody's riffing off each other,
nobody is stopping for a moment. -
9:31 - 9:34And so I tried to do the same
things, this time with music, -
9:34 - 9:37where the robot doesn't really know
what it's about to play, -
9:37 - 9:40it just sort of moves its body
and uses opportunities to play, -
9:40 - 9:42and does what my jazz teacher
when I was 17 taught me. -
9:42 - 9:44She said, when you improvise,
-
9:44 - 9:47sometimes you don't know
what you're doing, and you still do it. -
9:47 - 9:51So I tried to make a robot that doesn't
actually know what it's doing, -
9:51 - 9:52but is still doing it.
-
9:52 - 9:55So let's look at a few seconds
from this performance, -
9:55 - 9:57where the robot listens
to the human musician -
9:57 - 9:58and improvises.
-
9:58 - 10:01And then, look how the human
musician also responds -
10:01 - 10:02to what the robot is doing
-
10:02 - 10:05and picking up from its behavior,
-
10:05 - 10:09and at some point can even be surprised
by what the robot came up with. -
10:09 - 10:11(Music)
-
10:48 - 10:50(Music ends)
-
10:51 - 10:57(Applause)
-
10:57 - 10:59Being a musician
is not just about making notes, -
10:59 - 11:02otherwise nobody
would ever go see a live show. -
11:02 - 11:04Musicians also communicate
with their bodies, -
11:04 - 11:06with other band members,
with the audience, -
11:06 - 11:08they use their bodies
to express the music. -
11:08 - 11:10And I thought, we already have
a robot musician on stage, -
11:10 - 11:13why not make it be
a full-fledged musician? -
11:13 - 11:15And I started designing
a socially expressive head -
11:16 - 11:17for the robot.
-
11:17 - 11:19The head doesn’t actually
touch the marimba, -
11:19 - 11:21it just expresses what the music is like.
-
11:21 - 11:23These are some napkin sketches
from a bar in Atlanta -
11:23 - 11:26that was dangerously
located exactly halfway -
11:26 - 11:28between my lab and my home.
-
11:28 - 11:32So I spent, I would say, on average,
three to four hours a day there. -
11:32 - 11:33I think.
-
11:33 - 11:35(Laughter)
-
11:35 - 11:38And I went back to my animation
tools and tried to figure out -
11:38 - 11:40not just what a robotic
musician would look like, -
11:40 - 11:43but especially what a robotic
musician would move like, -
11:43 - 11:46to sort of show that it doesn't like
what the other person is playing -- -
11:46 - 11:50and maybe show whatever beat
it's feeling at the moment. -
11:50 - 11:54So we ended up actually getting the money
to build this robot, which was nice. -
11:54 - 11:57I'm going to show you now
the same kind of performance, -
11:57 - 11:59this time with a socially expressive head.
-
11:59 - 12:00And notice one thing --
-
12:00 - 12:02how the robot is really showing us
-
12:02 - 12:04the beat it's picking up from the human,
-
12:04 - 12:08while also giving the human a sense
that the robot knows what it's doing. -
12:08 - 12:10And also how it changes the way it moves
-
12:10 - 12:12as soon as it starts its own solo.
-
12:12 - 12:16(Music)
-
12:40 - 12:43Now look at the final chord
of the piece again. -
12:43 - 12:46And this time the robot
communicates with its body -
12:46 - 12:48when it's busy doing its own thing,
-
12:48 - 12:53and when it's ready to coordinate
the final chord with me. -
12:53 - 12:56(Music)
-
13:01 - 13:03(Music ending)
-
13:05 - 13:06(Final chord)
-
13:06 - 13:11(Applause)
-
13:11 - 13:12Thanks.
-
13:12 - 13:14I hope you see
-
13:14 - 13:18how much this part of the body
that doesn't touch the instrument -
13:19 - 13:21actually helps
with the musical performance. -
13:22 - 13:24And at some point -- we are in Atlanta,
-
13:24 - 13:27so obviously some rapper
will come into our lab at some point -- -
13:27 - 13:32and we had this rapper come in and do
a little jam with the robot. -
13:32 - 13:36Here you can see the robot
basically responding to the beat. -
13:36 - 13:37Notice two things:
-
13:37 - 13:41one, how irresistible it is to join
the robot while it's moving its head. -
13:41 - 13:43You kind of want to move
your own head when it does it. -
13:43 - 13:47And second, even though the rapper
is really focused on his iPhone, -
13:47 - 13:50as soon as the robot turns
to him, he turns back. -
13:50 - 13:53So even though it's just
in the periphery of his vision, -
13:53 - 13:55in the corner of his eye,
it's very powerful. -
13:55 - 13:57And the reason is that we can't ignore
-
13:57 - 13:59physical things moving in our environment.
-
13:59 - 14:00We are wired for that.
-
14:00 - 14:01So if you have a problem --
-
14:02 - 14:05maybe your partner is looking
at their iPhone or smartphone too much -- -
14:05 - 14:08you might want to have a robot
there to get their attention. -
14:08 - 14:09(Laughter)
-
14:09 - 14:11(Music)
-
14:24 - 14:26(Music ends)
-
14:26 - 14:33(Applause)
-
14:34 - 14:37Just to introduce the last robot
that we've worked on, -
14:37 - 14:40it came out of something
surprising that we found: -
14:40 - 14:43At some point people didn't care
about the robot being intelligent, -
14:43 - 14:44able to improvise and listen,
-
14:44 - 14:49and do all these embodied intelligence
things that I spent years developing. -
14:49 - 14:52They really liked that the robot
was enjoying the music. -
14:52 - 14:53(Laughter)
-
14:53 - 14:55And they didn't say the robot
was moving to the music, -
14:55 - 14:57they said "enjoying" the music.
-
14:57 - 14:59And we thought,
why don't we take this idea, -
14:59 - 15:02and I designed a new piece of furniture.
-
15:02 - 15:04This time it wasn't a desk lamp,
it was a speaker dock, -
15:04 - 15:07one of those things
you plug your smartphone in. -
15:07 - 15:08And I thought,
-
15:08 - 15:12what would happen if your speaker dock
didn't just play the music for you, -
15:12 - 15:14but would actually enjoy it, too?
-
15:14 - 15:18And so again, here are some
animation tests from an early stage. -
15:18 - 15:19(Laughter)
-
15:20 - 15:23And this is what the final
product looked like. -
15:34 - 15:36(Music)
-
15:56 - 15:58(Music ends)
-
15:59 - 16:01So, a lot of bobbing heads.
-
16:01 - 16:05(Applause)
-
16:07 - 16:09A lot of bobbing heads in the audience,
-
16:09 - 16:12so we can still see
robots influence people. -
16:12 - 16:15And it's not just fun and games.
-
16:15 - 16:17I think one of the reasons I care so much
-
16:17 - 16:19about robots that use
their body to communicate -
16:19 - 16:21and use their body to move is --
-
16:21 - 16:24I'm going to let you in on a little
secret we roboticists are hiding -- -
16:24 - 16:27is that every one of you
is going to be living with a robot -
16:27 - 16:29at some point in your life.
-
16:29 - 16:32Somewhere in your future,
there will be a robot in your life. -
16:32 - 16:33If not in yours, your children's lives.
-
16:34 - 16:38And I want these robots to be more fluent,
more engaging, more graceful -
16:39 - 16:40than currently they seem to be.
-
16:40 - 16:44And for that I think maybe robots
need to be less like chess players -
16:44 - 16:46and more like stage actors
and more like musicians. -
16:46 - 16:49Maybe they should be able
to take chances and improvise. -
16:49 - 16:52Maybe they should be able
to anticipate what you're about to do. -
16:52 - 16:55Maybe they even need to be able
to make mistakes and correct them, -
16:56 - 16:57because in the end, we are human.
-
16:57 - 17:01And maybe as humans,
robots that are a little less than perfect -
17:01 - 17:03are just perfect for us.
-
17:03 - 17:04Thank you.
-
17:04 - 17:11(Applause)
- Title:
- Robots with "soul" | Guy Hoffman | TEDxJaffa
- Description:
-
What kind of robots does an animator / jazz musician / roboticist make? Playful, reactive, curious ones. Guy Hoffman shows demo film of his family of unusual robots -- including two musical bots that like to jam with humans.
This talk was given at a TEDx event using the TED conference format but independently organized by a local community. Learn more at http://ted.com/tedx
- Video Language:
- English
- Team:
- closed TED
- Project:
- TEDxTalks
- Duration:
- 17:26
Ivana Korom approved English subtitles for Robots with "soul" | Guy Hoffman | TEDxJaffa | ||
Ivana Korom accepted English subtitles for Robots with "soul" | Guy Hoffman | TEDxJaffa | ||
TED Translators admin edited English subtitles for Robots with "soul" | Guy Hoffman | TEDxJaffa | ||
TED Translators admin edited English subtitles for Robots with "soul" | Guy Hoffman | TEDxJaffa | ||
TED Translators admin edited English subtitles for Robots with "soul" | Guy Hoffman | TEDxJaffa | ||
TED Translators admin edited English subtitles for Robots with "soul" | Guy Hoffman | TEDxJaffa | ||
TED Translators admin edited English subtitles for Robots with "soul" | Guy Hoffman | TEDxJaffa | ||
TED Translators admin edited English subtitles for Robots with "soul" | Guy Hoffman | TEDxJaffa |