TED2007
Hod Lipson: Building "self-aware" robots
Hod Lipson izrađuje samosvjesne robote
Filmed:
Readability: 3.1
1,460,460 views
Hod Lipson demostrira nekoliko svojih cool malih robota, koji imaju sposobnost učiti, razumjeti sami sebe i čak replicirati sami sebe.
Hod Lipson - Roboticist
Hod Lipson works at the intersection of engineering and biology, studying robots and the way they "behave" and evolve. His work has exciting implications for design and manufacturing -- and serves as a window to understand our own behavior and evolution. Full bio
Hod Lipson works at the intersection of engineering and biology, studying robots and the way they "behave" and evolve. His work has exciting implications for design and manufacturing -- and serves as a window to understand our own behavior and evolution. Full bio
Double-click the English transcript below to play the video.
00:25
So, where are the robots?
0
0
2000
Dakle, što su roboti?
00:27
We've been told for 40 years already that they're coming soon.
1
2000
3000
Već nam se 40 godina priča kako će se uskoro pojaviti.
00:30
Very soon they'll be doing everything for us.
2
5000
3000
Ubrzo oni će raditi sve za nas:
00:33
They'll be cooking, cleaning, buying things, shopping, building. But they aren't here.
3
8000
5000
oni će kuhati, čistiti, kupovati stvari, ići u kupovinu, graditi. Ali oni nisu ovdje.
00:38
Meanwhile, we have illegal immigrants doing all the work,
4
13000
4000
U međuvremenu, imamo ilegalne imigrante koji rade sav taj posao,
00:42
but we don't have any robots.
5
17000
2000
ali nemamo robote.
00:44
So what can we do about that? What can we say?
6
19000
4000
Dakle, što možemo učiniti oko toga? Što možemo reći?
00:48
So I want to give a little bit of a different perspective
7
23000
4000
Dakle, želim dati malo drugačiju perspektivu
00:52
of how we can perhaps look at these things in a little bit of a different way.
8
27000
6000
o tome kako možda možemo gledati na te stvari na pomalo drugačiji način.
00:58
And this is an x-ray picture
9
33000
2000
A ovo je rendgenska snimka
01:00
of a real beetle, and a Swiss watch, back from '88. You look at that --
10
35000
5000
prave bube i švicarskog sata, iz 1988.godine. Gledate u to --
01:05
what was true then is certainly true today.
11
40000
2000
ono što je bilo istinito tada je definitivno istinito danas.
01:07
We can still make the pieces. We can make the right pieces.
12
42000
3000
Mi još uvijek možemo izraditi dijelove, mi možemo izraditi prave dijelove,
01:10
We can make the circuitry of the right computational power,
13
45000
3000
mi možemo izraditi strujne krugove prave računalne moći,
01:13
but we can't actually put them together to make something
14
48000
3000
ali mi ih zapravo ne možemo staviti zajedno kako bi napravili nešto
01:16
that will actually work and be as adaptive as these systems.
15
51000
5000
što će zapravo raditi i biti prilagodljivo poput ovih sustava.
01:21
So let's try to look at it from a different perspective.
16
56000
2000
Dakle, pokušajmo ih gledati iz druge perspektive.
01:23
Let's summon the best designer, the mother of all designers.
17
58000
4000
Pozovimo najboljeg dizajnera, majku svih dizajnera:
01:27
Let's see what evolution can do for us.
18
62000
3000
da vidimo što evolucija može učiniti za nas.
01:30
So we threw in -- we created a primordial soup
19
65000
4000
Dakle, ubacili smo -- stvorili smo praiskonsku juhu
01:34
with lots of pieces of robots -- with bars, with motors, with neurons.
20
69000
4000
s mnogim dijelovima robota: sa šipkama, s motorima, s neuronima.
01:38
Put them all together, and put all this under kind of natural selection,
21
73000
4000
Stavite ih sve zajedno, i stavite sve to pod jednu vrstu prirodne selekcije,
01:42
under mutation, and rewarded things for how well they can move forward.
22
77000
4000
pod mutaciju, i nagrađene stvari ovisno o tome kako dobro se mogu kretati naprijed.
01:46
A very simple task, and it's interesting to see what kind of things came out of that.
23
81000
6000
Jako jednostavan zadatak, i zanimljivo je vidjeti kakve vrste stvari su proizašle iz toga.
01:52
So if you look, you can see a lot of different machines
24
87000
3000
Ako pogledate, možete vidjeti kako je mnogo različitih strojeva
01:55
come out of this. They all move around.
25
90000
2000
proizašlo iz toga. Svi se oni kreću okolo,
01:57
They all crawl in different ways, and you can see on the right,
26
92000
4000
svi oni pužu u različitim smjerovima, i možete vidjeti ovdje na desnoj strani,
02:01
that we actually made a couple of these things,
27
96000
2000
da smo zapravo napravili par tih stvari,
02:03
and they work in reality. These are not very fantastic robots,
28
98000
3000
i one rade u stvarnosti. Ovo nisu jako fantastični roboti,
02:06
but they evolved to do exactly what we reward them for:
29
101000
4000
ali su evoluirali do toga da rade točno ono za što ih nagrađujemo:
02:10
for moving forward. So that was all done in simulation,
30
105000
3000
za kretanje unaprijed. Dakle, to je sve napravljeno pomoću simulacije,
02:13
but we can also do that on a real machine.
31
108000
2000
ali možemo to učiniti i na pravom stroju.
02:15
Here's a physical robot that we actually
32
110000
5000
Ovdje je fizički robot kojem smo zapravo
02:20
have a population of brains,
33
115000
3000
usadili mozak,
02:23
competing, or evolving on the machine.
34
118000
2000
koji se takmiči, ili evoluira, na stroju.
02:25
It's like a rodeo show. They all get a ride on the machine,
35
120000
3000
To je poput rodea: svi oni dobiju vožnju na stroju,
02:28
and they get rewarded for how fast or how far
36
123000
3000
i svi oni budu nagrađeni ovisno kako brzo ili kako daleko
02:31
they can make the machine move forward.
37
126000
2000
mogu natjerati stroj da se kreće naprijed.
02:33
And you can see these robots are not ready
38
128000
2000
I možete vidjeti da ti roboti nisu spremni
02:35
to take over the world yet, but
39
130000
3000
da preuzmu svijet još, ali
02:38
they gradually learn how to move forward,
40
133000
2000
oni postepeno uče kako se kretati naprijed,
02:40
and they do this autonomously.
41
135000
3000
i to rade samostalno.
02:43
So in these two examples, we had basically
42
138000
4000
Dakle, u ova dva primjera, mi smo u osnovi imali
02:47
machines that learned how to walk in simulation,
43
142000
3000
strojeve koji su učili kako hodati u simulaciji,
02:50
and also machines that learned how to walk in reality.
44
145000
2000
i isto tako strojeve koji su učili kako hodati u stvarnosti.
02:52
But I want to show you a different approach,
45
147000
2000
Ali želim vam pokazati drugačiji pristup,
02:54
and this is this robot over here, which has four legs.
46
149000
6000
a to je ovaj robot, ovdje, koji ima četiri noge,
03:00
It has eight motors, four on the knees and four on the hip.
47
155000
2000
ima osam motora, četiri na koljenima i četiri na boku.
03:02
It has also two tilt sensors that tell the machine
48
157000
3000
Ujedno ima i dva nagibna senzora koji govore stroju
03:05
which way it's tilting.
49
160000
3000
u kojem smjeru je nagib.
03:08
But this machine doesn't know what it looks like.
50
163000
2000
Ali ovaj stroj ne zna kako izgleda.
03:10
You look at it and you see it has four legs,
51
165000
2000
Gledate u njega i vidite da ima četiri noge,
03:12
the machine doesn't know if it's a snake, if it's a tree,
52
167000
2000
stroj ne zna je li to zmija, ili drvo,
03:14
it doesn't have any idea what it looks like,
53
169000
3000
nema nikakvu ideju o tome kako izgleda,
03:17
but it's going to try to find that out.
54
172000
2000
ali pokušat će to doznati.
03:19
Initially, it does some random motion,
55
174000
2000
Inicijalno, radi neke nasumične pokrete,
03:21
and then it tries to figure out what it might look like.
56
176000
3000
i zatim pokušava dokučiti na što bi mogao ličiti --
03:24
And you're seeing a lot of things passing through its minds,
57
179000
2000
i vidite mnogo stvari koje prolaze kroz njegove misli,
03:26
a lot of self-models that try to explain the relationship
58
181000
4000
mnogo samo-modela koji pokušavaju objasniti vezu
03:30
between actuation and sensing. It then tries to do
59
185000
3000
između stavljanja u pokret i osjećanja -- i zatim pokušava uraditi
03:33
a second action that creates the most disagreement
60
188000
4000
drugu radnju koja stvara najviše neslaganja
03:37
among predictions of these alternative models,
61
192000
2000
između predviđanja tih opcijskih modela,
03:39
like a scientist in a lab. Then it does that
62
194000
2000
poput znanstvenika u laboratoriju. Zatim radi to
03:41
and tries to explain that, and prune out its self-models.
63
196000
4000
i pokušava to objasniti, i izrezati vlastite samo-modele.
03:45
This is the last cycle, and you can see it's pretty much
64
200000
3000
Ovo je posljednji ciklus, i možete vidjeti da je više-manje
03:48
figured out what its self looks like. And once it has a self-model,
65
203000
4000
dokučio kako njegovo biće izgleda, i jednom kada ima samo-model,
03:52
it can use that to derive a pattern of locomotion.
66
207000
4000
može to iskoristiti da izvuće uzorak kretanja.
03:56
So what you're seeing here are a couple of machines --
67
211000
2000
Dakle, ono što vidite ovdje je par strojeva --
03:58
a pattern of locomotion.
68
213000
2000
uzorak kretanja.
04:00
We were hoping that it wass going to have a kind of evil, spidery walk,
69
215000
4000
Nadali smo se kako će imati tu neku vrstu zlobnog, paukovskog hoda,
04:04
but instead it created this pretty lame way of moving forward.
70
219000
4000
ali umjesto toga, stvorio je prilično jadan način kretanja naprijed.
04:08
But when you look at that, you have to remember
71
223000
3000
Ali kada to gledate, morate upamtiti
04:11
that this machine did not do any physical trials on how to move forward,
72
226000
6000
kako taj stroj nije radio nikakve fizičke pokuse kako se kretati unaprijed,
04:17
nor did it have a model of itself.
73
232000
2000
niti je imao model samog sebe.
04:19
It kind of figured out what it looks like, and how to move forward,
74
234000
3000
Nekako je sam dokučio kako izgleda, i kako se treba kretati naprijed,
04:22
and then actually tried that out.
75
237000
4000
i zatim je zapravo to pokušao.
04:26
(Applause)
76
241000
5000
(Pljesak)
04:31
So, we'll move forward to a different idea.
77
246000
4000
Dakle, mi ćemo krenuti na drugačiju ideju.
04:35
So that was what happened when we had a couple of --
78
250000
5000
Dakle, to se dogodilo kada smo imali par --
04:40
that's what happened when you had a couple of -- OK, OK, OK --
79
255000
4000
to se dogodilo kada si imao par -- OK, OK, OK --
04:44
(Laughter)
80
259000
2000
(Smijeh)
04:46
-- they don't like each other. So
81
261000
2000
-- ne vole jedan drugoga. Dakle,
04:48
there's a different robot.
82
263000
3000
ovdje je različit robot.
04:51
That's what happened when the robots actually
83
266000
2000
To je ono što se dogodilo kada roboti zapravo
04:53
are rewarded for doing something.
84
268000
2000
budu nagrađeni za nešto što su napravili.
04:55
What happens if you don't reward them for anything, you just throw them in?
85
270000
3000
Što se događa ako ih ne nagradite za bilošto, već ih samo ubacite unutra?
04:58
So we have these cubes, like the diagram showed here.
86
273000
3000
Dakle, imamo te kocke, kako je to ovdje prikazano na dijagramu.
05:01
The cube can swivel, or flip on its side,
87
276000
2000
Kocka se može okretati, ili se preokrenuti na stranu,
05:04
and we just throw 1,000 of these cubes into a soup --
88
279000
4000
i mi samo ubacimo 1.000 takvih kocki u juhu --
05:08
this is in simulation --and don't reward them for anything,
89
283000
2000
ovo je u simulaciji -- i ne nagradimo ih za išta,
05:10
we just let them flip. We pump energy into this
90
285000
3000
jednostavno im damo da polude. Upumpavamo energiju u to
05:13
and see what happens in a couple of mutations.
91
288000
3000
i vidimo što se događa u par mutacija.
05:16
So, initially nothing happens, they're just flipping around there.
92
291000
3000
Dakle, inicijalno, ništa se ne događa, samo lude tamo.
05:19
But after a very short while, you can see these blue things
93
294000
4000
Ali nedugo nakon toga, možete vidjeti ove plave stvari
05:23
on the right there begin to take over.
94
298000
2000
na desnoj strani koje počinju preuzimati.
05:25
They begin to self-replicate. So in absence of any reward,
95
300000
4000
Počinju se samo-razmnožavati. Dakle, u odsustvu ikakve nagrade,
05:29
the intrinsic reward is self-replication.
96
304000
3000
intrinzična nagrada je samo-razmnožavanje.
05:32
And we've actually built a couple of these,
97
307000
1000
I mi smo zapravo izradili par tih,
05:33
and this is part of a larger robot made out of these cubes.
98
308000
4000
i to je dio većeg robota koji je napravljen od tih kocki,
05:37
It's an accelerated view, where you can see the robot actually
99
312000
3000
to je ubrzan prikaz, gdje možete vidjeti kako robot zapravo
05:40
carrying out some of its replication process.
100
315000
2000
prolazi kroz neki od procesa razmnožavanja.
05:42
So you're feeding it with more material -- cubes in this case --
101
317000
4000
Dakle, hranite ga s više materijala -- kocki u ovom slučaju --
05:46
and more energy, and it can make another robot.
102
321000
3000
i više energije, i može stvoriti još jedan robot.
05:49
So of course, this is a very crude machine,
103
324000
3000
Dakle, naravno, ovo je vrlo sirov stroj,
05:52
but we're working on a micro-scale version of these,
104
327000
2000
ali radimo na mikro-verziji njih,
05:54
and hopefully the cubes will be like a powder that you pour in.
105
329000
3000
i nadamo se kako će te kocke biti poput praha koji ulijete.
05:57
OK, so what can we learn? These robots are of course
106
332000
5000
U redu, dakle, što možemo naučiti? Ti roboti naravno
06:02
not very useful in themselves, but they might teach us something
107
337000
3000
nisu sami po sebi korisni, ali bi nas mogli naučiti nešto
06:05
about how we can build better robots,
108
340000
3000
o tome kako možemo izraditi bolje robote,
06:08
and perhaps how humans, animals, create self-models and learn.
109
343000
5000
i možda kako ljudi, životinje, stvaraju samo-modele i uče.
06:13
And one of the things that I think is important
110
348000
2000
I jedna od stvari za koju smatram da je važna
06:15
is that we have to get away from this idea
111
350000
2000
da se moramo odmaknuti od te ideje
06:17
of designing the machines manually,
112
352000
2000
ručnog dizajniranja strojeva,
06:19
but actually let them evolve and learn, like children,
113
354000
3000
već im dopustiti da evoluiraju i uče, poput djece,
06:22
and perhaps that's the way we'll get there. Thank you.
114
357000
2000
i možda je to način da stignemo tamo. Hvala vam.
06:24
(Applause)
115
359000
2000
(Pljesak)
ABOUT THE SPEAKER
Hod Lipson - RoboticistHod Lipson works at the intersection of engineering and biology, studying robots and the way they "behave" and evolve. His work has exciting implications for design and manufacturing -- and serves as a window to understand our own behavior and evolution.
Why you should listen
To say that Hod Lipson and his team at Cornell build robots is not completely accurate: They may simply set out a pile of virtual robot parts, devise some rules for assembly, and see what the parts build themselves into. They've created robots that decide for themselves how they want to walk; robots that develop a sense of what they look like; even robots that can, through trial and error, construct other robots just like themselves.
Working across disciplines -- physics, computer science, math, biology and several flavors of engineer -- the team studies techniques for self-assembly and evolution that have great implications for fields such as micro-manufacturing -- allowing tiny pieces to assemble themselves at scales heretofore impossible -- and extreme custom manufacturing (in other words, 3-D printers for the home).
His lab's Outreach page is a funhouse of tools and instructions, including the amazing Golem@Home -- a self-assembling virtual robot who lives in your screensaver.
Hod Lipson | Speaker | TED.com