TED2007
Hod Lipson: Building "self-aware" robots
Hod Lipson stavia robotov, ktorí si uvedomujú sami seba
Filmed:
Readability: 3.1
1,460,460 views
Hod Lipson ukazuje zopár svojich úžasných malých robotov, ktorí majú schopnosť učiť sa, chápať sami seba a dokonca sa sami replikovať.
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
Takže, kde sú tí roboti?
00:27
We've been told for 40 years already that they're coming soon.
1
2000
3000
Už 40 rokov nám hovoria, že čoskoro prídu.
00:30
Very soon they'll be doing everything for us.
2
5000
3000
Čoskoro budú pre nás robiť čokoľvek:
00:33
They'll be cooking, cleaning, buying things, shopping, building. But they aren't here.
3
8000
5000
variť, upratovať, kupovať veci, nakupovať, stavať. Ale nie sú tu.
00:38
Meanwhile, we have illegal immigrants doing all the work,
4
13000
4000
Na prácu máme nelegálnych prisťahovalcov,
00:42
but we don't have any robots.
5
17000
2000
ale žiadnych robotov.
00:44
So what can we do about that? What can we say?
6
19000
4000
Čo s tým môžeme urobiť? Čo môžeme povedať?
00:48
So I want to give a little bit of a different perspective
7
23000
4000
Chcel by som vám ukázať trochu iný uhol pohľadu na to,
00:52
of how we can perhaps look at these things in a little bit of a different way.
8
27000
6000
ako by sme sa mohli pozrieť na tieto veci.
00:58
And this is an x-ray picture
9
33000
2000
Toto je rontgenový snímok
01:00
of a real beetle, and a Swiss watch, back from '88. You look at that --
10
35000
5000
skutočného chrobáka a hodiniek Swiss z roku 88. Pozrite sa na to --
01:05
what was true then is certainly true today.
11
40000
2000
čo bola pravda vtedy je celkom určite pravda aj dnes.
01:07
We can still make the pieces. We can make the right pieces.
12
42000
3000
Vieme vyrobiť časti, vieme vyrobiť správne časti,
01:10
We can make the circuitry of the right computational power,
13
45000
3000
vieme vyrobiť obvody správnej výpočtovej sily,
01:13
but we can't actually put them together to make something
14
48000
3000
ale nevieme ich poskladať dokopy a vyrobiť niečo,
01:16
that will actually work and be as adaptive as these systems.
15
51000
5000
čo by naozaj fungovalo a vedelo sa prispôsobiť ako tieto systémy.
01:21
So let's try to look at it from a different perspective.
16
56000
2000
Skúsme sa na to pozrieť z iného uhla pohľadu.
01:23
Let's summon the best designer, the mother of all designers.
17
58000
4000
Inšpirujme sa najlepšou návrhárkou, matkou všetkých návrhárov,
01:27
Let's see what evolution can do for us.
18
62000
3000
pozrime sa, čo pre nás môže urobiť evolúcia.
01:30
So we threw in -- we created a primordial soup
19
65000
4000
Tak sme sa do toho pustili. Vytvorili sme prvotnú kopu
01:34
with lots of pieces of robots -- with bars, with motors, with neurons.
20
69000
4000
zo všetkých častí robotov - z tyčí, z motorov, z neurónov.
01:38
Put them all together, and put all this under kind of natural selection,
21
73000
4000
Dali sme ich ku sebe, pustili sme na to prirodzený výber
01:42
under mutation, and rewarded things for how well they can move forward.
22
77000
4000
pomocou mutácie a odmeňovali sme veci podľa toho, ako sa vedeli posúvať dopredu.
01:46
A very simple task, and it's interesting to see what kind of things came out of that.
23
81000
6000
Veľmi jednoduchá úloha a je zaujímavé, čo z toho vzišlo.
01:52
So if you look, you can see a lot of different machines
24
87000
3000
Ak sa pozriete, môžete vidieť množstvo rôznych strojov,
01:55
come out of this. They all move around.
25
90000
2000
ktoré sa takto vytvorili. Pohybujú sa naokolo,
01:57
They all crawl in different ways, and you can see on the right,
26
92000
4000
plazia sa rôznymi spôsobmi a napravo vidíte,
02:01
that we actually made a couple of these things,
27
96000
2000
že sme vyrobili niekoľko z týchto vecí
02:03
and they work in reality. These are not very fantastic robots,
28
98000
3000
a fungujú aj v reálnom svete. Nie sú to fantastické roboty,
02:06
but they evolved to do exactly what we reward them for:
29
101000
4000
ale boli vyvinuté tak, aby robili presne to, za čo sme ich odmeňovali:
02:10
for moving forward. So that was all done in simulation,
30
105000
3000
pohybovali sa dopredu. Toto všetko sa dialo v simulácii,
02:13
but we can also do that on a real machine.
31
108000
2000
ale vieme to urobiť aj na skutočnom stroji.
02:15
Here's a physical robot that we actually
32
110000
5000
Tu je skutočný robot, ktorý má
02:20
have a population of brains,
33
115000
3000
niekoľko mozgov, ktoré
02:23
competing, or evolving on the machine.
34
118000
2000
v tomto stroji súťažia a vyvíjajú sa.
02:25
It's like a rodeo show. They all get a ride on the machine,
35
120000
3000
Je to ako rodeo show, striedajú sa v tom, kto jazdí na stroji
02:28
and they get rewarded for how fast or how far
36
123000
3000
a dostávajú odmenu podľa toho, ako rýchlo alebo ako ďaleko
02:31
they can make the machine move forward.
37
126000
2000
s ním zájdu.
02:33
And you can see these robots are not ready
38
128000
2000
A hoci títo roboti nie sú pripravení
02:35
to take over the world yet, but
39
130000
3000
prevziať vládu nad svetom,
02:38
they gradually learn how to move forward,
40
133000
2000
postupne sa učia, ako sa posúvať dopredu
02:40
and they do this autonomously.
41
135000
3000
a učia sa samostatne.
02:43
So in these two examples, we had basically
42
138000
4000
V týchto dvoch príkladoch sme mali
02:47
machines that learned how to walk in simulation,
43
142000
3000
stroje, ktoré sa naučili chodiť v simulácii
02:50
and also machines that learned how to walk in reality.
44
145000
2000
a tiež stroje, ktoré sa naučili chodiť v reálnom svete.
02:52
But I want to show you a different approach,
45
147000
2000
Ale chcel by som vám ukázať iný prístup,
02:54
and this is this robot over here, which has four legs.
46
149000
6000
a to tohto robota, ktorý má štyri nohy,
03:00
It has eight motors, four on the knees and four on the hip.
47
155000
2000
osem motorov, štyri na kolenách a štyri na bokoch.
03:02
It has also two tilt sensors that tell the machine
48
157000
3000
Tiež má dva senzory sklonu, ktoré mu hovoria,
03:05
which way it's tilting.
49
160000
3000
na ktorú stranu je naklonený.
03:08
But this machine doesn't know what it looks like.
50
163000
2000
Ale tento stroj nevie, ako vyzerá.
03:10
You look at it and you see it has four legs,
51
165000
2000
Vy sa naňho pozriete a vidíte, že má štyri nohy,
03:12
the machine doesn't know if it's a snake, if it's a tree,
52
167000
2000
ale on nevie, či vyzerá ako had alebo strom,
03:14
it doesn't have any idea what it looks like,
53
169000
3000
nemá predstavu o tom, ako vyzerá,
03:17
but it's going to try to find that out.
54
172000
2000
ale skúsi to zistiť.
03:19
Initially, it does some random motion,
55
174000
2000
Najprv robí náhodné pohyby,
03:21
and then it tries to figure out what it might look like.
56
176000
3000
a potom sa snaží zistiť, ako by mohol vyzerať --
03:24
And you're seeing a lot of things passing through its minds,
57
179000
2000
a vidíte, že mu napadá veľa vecí,
03:26
a lot of self-models that try to explain the relationship
58
181000
4000
veľa modelov seba samého, ktoré sa snažia vysvetliť vzťah
03:30
between actuation and sensing. It then tries to do
59
185000
3000
medzi podnetom a vnemom -- potom sa snaží urobiť
03:33
a second action that creates the most disagreement
60
188000
4000
druhú akciu, ktorá vytvára najväčšiu nezhodu
03:37
among predictions of these alternative models,
61
192000
2000
medzi predpoveďami týchto alternatívnych modelov,
03:39
like a scientist in a lab. Then it does that
62
194000
2000
tak ako vedec v laboratóriu. Potom to urobí
03:41
and tries to explain that, and prune out its self-models.
63
196000
4000
a snaží sa to vysvetliť, a postupne vyraďuje modely.
03:45
This is the last cycle, and you can see it's pretty much
64
200000
3000
Toto je posledný cyklus a môžete vidieť, že celkom dobre
03:48
figured out what its self looks like. And once it has a self-model,
65
203000
4000
uhádol, ako vyzerá, a keď už má model seba samého,
03:52
it can use that to derive a pattern of locomotion.
66
207000
4000
môže ho použiť na odvodenie systému premiestňovania.
03:56
So what you're seeing here are a couple of machines --
67
211000
2000
Tu vidíte niekoľko strojov --
03:58
a pattern of locomotion.
68
213000
2000
systém premiestňovania.
04:00
We were hoping that it wass going to have a kind of evil, spidery walk,
69
215000
4000
Dúfali sme, že bude mať nejakú diabolskú, pavúčiu chôdzu,
04:04
but instead it created this pretty lame way of moving forward.
70
219000
4000
ale namiesto toho vytvoril tento trochu trápny spôsob posúvania sa dopredu.
04:08
But when you look at that, you have to remember
71
223000
3000
Ale keď sa na to pozeráte, nesmiete zabudnúť,
04:11
that this machine did not do any physical trials on how to move forward,
72
226000
6000
že tento stroj nikdy predtým neskúšal, ako sa posúvať dopredu,
04:17
nor did it have a model of itself.
73
232000
2000
ani nemal model seba samého.
04:19
It kind of figured out what it looks like, and how to move forward,
74
234000
3000
Sám zistil, ako vyzerá, a ako sa má posúvať,
04:22
and then actually tried that out.
75
237000
4000
a až potom to skúsil.
04:26
(Applause)
76
241000
5000
(Potlesk)
04:31
So, we'll move forward to a different idea.
77
246000
4000
Tak, presuňme sa ďalej.
04:35
So that was what happened when we had a couple of --
78
250000
5000
Toto sa stalo, keď sme mali pár --
04:40
that's what happened when you had a couple of -- OK, OK, OK --
79
255000
4000
toto sa stalo, keď sme mali pár -- OK, OK, OK --
04:44
(Laughter)
80
259000
2000
(Smiech)
04:46
-- they don't like each other. So
81
261000
2000
-- nemajú sa radi. Takže,
04:48
there's a different robot.
82
263000
3000
toto je iný robot.
04:51
That's what happened when the robots actually
83
266000
2000
Toto sa stane, keď sú roboti
04:53
are rewarded for doing something.
84
268000
2000
odmeňovaní za to, že niečo robia.
04:55
What happens if you don't reward them for anything, you just throw them in?
85
270000
3000
Čo sa stane, keď ich za nič neodmeníte, keď ich len tak necháte?
04:58
So we have these cubes, like the diagram showed here.
86
273000
3000
Mali sme tieto kocky, ako vidíte na diagrame.
05:01
The cube can swivel, or flip on its side,
87
276000
2000
Kocka sa môže otáčať alebo sa prevrátiť na stranu.
05:04
and we just throw 1,000 of these cubes into a soup --
88
279000
4000
My sme dali 1000 týchto kociek na kopu --
05:08
this is in simulation --and don't reward them for anything,
89
283000
2000
toto je simulácia -- a za nič sme ich neodmeňovali,
05:10
we just let them flip. We pump energy into this
90
285000
3000
len sme ich nechali prevracať sa. Napumpovali sme do toho energiu
05:13
and see what happens in a couple of mutations.
91
288000
3000
a čakali, čo sa stane po niekoľkých mutáciách.
05:16
So, initially nothing happens, they're just flipping around there.
92
291000
3000
Na začiatku sa nič nedialo, len sa tam prevracali.
05:19
But after a very short while, you can see these blue things
93
294000
4000
Ale po krátkom čase, môžete vidieť, že tieto modré veci
05:23
on the right there begin to take over.
94
298000
2000
napravo začínajú viesť.
05:25
They begin to self-replicate. So in absence of any reward,
95
300000
4000
Začali sa replikovať. Takže, ak chýba odmena,
05:29
the intrinsic reward is self-replication.
96
304000
3000
vrodenou odmenou je replikovanie.
05:32
And we've actually built a couple of these,
97
307000
1000
Vytvorili sme zopár takýchto robotov
05:33
and this is part of a larger robot made out of these cubes.
98
308000
4000
a toto je časť väčšieho robota vytvoreného z týchto kociek.
05:37
It's an accelerated view, where you can see the robot actually
99
312000
3000
V tomto zrýchlenom pohľade môžete vidieť, že robot
05:40
carrying out some of its replication process.
100
315000
2000
vykonáva svoj proces reprodukcie.
05:42
So you're feeding it with more material -- cubes in this case --
101
317000
4000
Takže ak mu dáte viac materiálu -- kociek v tomto prípade --
05:46
and more energy, and it can make another robot.
102
321000
3000
a viac energie, môže urobiť ďalšieho robota.
05:49
So of course, this is a very crude machine,
103
324000
3000
Samozrejme, toto je veľmi hrubý návrh stroja,
05:52
but we're working on a micro-scale version of these,
104
327000
2000
ale pracujeme na mikroverzii takýchto,
05:54
and hopefully the cubes will be like a powder that you pour in.
105
329000
3000
a snáď tieto kocky budú vo forme prášku, ktorý tam vsypete.
05:57
OK, so what can we learn? These robots are of course
106
332000
5000
Takže, čo sa z tohto môžeme naučiť? Títo roboti nie sú
06:02
not very useful in themselves, but they might teach us something
107
337000
3000
veľmi užitoční sami o sebe, ale môžu nás naučiť niečo o tom,
06:05
about how we can build better robots,
108
340000
3000
ako budovať lepších robotov,
06:08
and perhaps how humans, animals, create self-models and learn.
109
343000
5000
a možno aj to, ako ľudia a zvieratá vytvárajú modely o sebe a učia sa.
06:13
And one of the things that I think is important
110
348000
2000
A jedna z vecí, ktorá je podľa mňa dôležitá, je,
06:15
is that we have to get away from this idea
111
350000
2000
že musíme upustiť od myšlienky
06:17
of designing the machines manually,
112
352000
2000
navrhovania strojov ručne,
06:19
but actually let them evolve and learn, like children,
113
354000
3000
ale namiesto toho im dovoliť vyvíjať sa a učiť sa, ako deti.
06:22
and perhaps that's the way we'll get there. Thank you.
114
357000
2000
Možno toto je spôsob, ako to dosiahnuť. Ďakujem.
06:24
(Applause)
115
359000
2000
(Potlesk)
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