TED2007
Hod Lipson: Building "self-aware" robots
Hod Lipson contruieste roboti "constienti de sine"
Filmed:
Readability: 3.1
1,460,460 views
Hod Lipson face o demonstratie cu roboteii lui, care sunt capabili sa invete, sa se intelega unul pe celalalt si chiar sa se autoreproduca.
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
Deci, unde sunt robotii?
00:27
We've been told for 40 years already that they're coming soon.
1
2000
3000
De 40 de ani ni se spune ca vor veni in curand.
00:30
Very soon they'll be doing everything for us.
2
5000
3000
In curand o sa faca totul in locul nostru:
00:33
They'll be cooking, cleaning, buying things, shopping, building. But they aren't here.
3
8000
5000
o sa gateasca,o sa faca curat,o sa faca cumparaturi, o sa construiasca.Dar nu sunt aici.
00:38
Meanwhile, we have illegal immigrants doing all the work,
4
13000
4000
Intre timp ,avem imigranti ilegali care fac toata treaba,
00:42
but we don't have any robots.
5
17000
2000
dar nu avem nici un robot.
00:44
So what can we do about that? What can we say?
6
19000
4000
Asa ca ce putem sa facem? Ce putem sa spunem?
00:48
So I want to give a little bit of a different perspective
7
23000
4000
As vrea sa va arat o modalitate alternativa
00:52
of how we can perhaps look at these things in a little bit of a different way.
8
27000
6000
despre cum ne putem uita la lucruri intr-un mod putin diferit.
00:58
And this is an x-ray picture
9
33000
2000
Aceasta este o radiografie
01:00
of a real beetle, and a Swiss watch, back from '88. You look at that --
10
35000
5000
a unui gandac adevarat, si a unui ceas elvetian, din '88.Te uiti la --
01:05
what was true then is certainly true today.
11
40000
2000
ce era adevarat atunci cu siguranta este adevarat si astazi.
01:07
We can still make the pieces. We can make the right pieces.
12
42000
3000
Inca mai putem sa fabricam piesele, putem face piesele corecte.
01:10
We can make the circuitry of the right computational power,
13
45000
3000
putem sa facem o placuta cu circuite de calcul,
01:13
but we can't actually put them together to make something
14
48000
3000
dar nu putem sa le punem la un loc sa facem ceva anume
01:16
that will actually work and be as adaptive as these systems.
15
51000
5000
care sa functioneze cu adevarat si sa fie capabil sa se adapteze la fel ca aceste sisteme.
01:21
So let's try to look at it from a different perspective.
16
56000
2000
Asa ca sa incercam sa privim lucrurile dintr-o alta perspectiva.
01:23
Let's summon the best designer, the mother of all designers.
17
58000
4000
Sa-l luam pe cel mai bun designer, cel mai bun designer dintre toti:
01:27
Let's see what evolution can do for us.
18
62000
3000
sa vedem ce poate face evolutia pentru noi.
01:30
So we threw in -- we created a primordial soup
19
65000
4000
Asa ca am amestecat-- am creat supa primordiala
01:34
with lots of pieces of robots -- with bars, with motors, with neurons.
20
69000
4000
cu multe bucati de roboti : cu fiare, cu motoare , cu neuroni.
01:38
Put them all together, and put all this under kind of natural selection,
21
73000
4000
Le adunam pe toate la un loc, si le supunem unui fel de proces natural de selectie,
01:42
under mutation, and rewarded things for how well they can move forward.
22
77000
4000
unui proces de transformare, si vedem cat de bine au sa evolueze.
01:46
A very simple task, and it's interesting to see what kind of things came out of that.
23
81000
6000
O sarcina foarte simpla, si e interesant de vazut ce fel de chestii rezulta.
01:52
So if you look, you can see a lot of different machines
24
87000
3000
Asa ca daca va uitati, o sa vedeti o gramada de masinarii diferite
01:55
come out of this. They all move around.
25
90000
2000
care au iesit din asta.Toate se misca,
01:57
They all crawl in different ways, and you can see on the right,
26
92000
4000
intr-un fel sau altul,puteti vedea in dreapta,
02:01
that we actually made a couple of these things,
27
96000
2000
chiar am creat niste chestii de astea,
02:03
and they work in reality. These are not very fantastic robots,
28
98000
3000
si chiar functioneaza. Nu sunt cine stie ce roboti,
02:06
but they evolved to do exactly what we reward them for:
29
101000
4000
dar au evoluat si au ajuns sa facea ce le-am cerut:
02:10
for moving forward. So that was all done in simulation,
30
105000
3000
se mearga inainte.Toate aceastea au fost facute intr-o simulare,
02:13
but we can also do that on a real machine.
31
108000
2000
dar putem face asta si cu o masinarie reala.
02:15
Here's a physical robot that we actually
32
110000
5000
Acesta este un robot pe care avem
02:20
have a population of brains,
33
115000
3000
o populatie de creiere,
02:23
competing, or evolving on the machine.
34
118000
2000
care concureaza unele cu celelalte, sau evoluaza, pe robot.
02:25
It's like a rodeo show. They all get a ride on the machine,
35
120000
3000
E ca la un rodeo show: toti apuca sa controleze masinaria,
02:28
and they get rewarded for how fast or how far
36
123000
3000
si sunt recompensati pentru cat de repede sau cat de departe
02:31
they can make the machine move forward.
37
126000
2000
au facut masinaria sa mearga.
02:33
And you can see these robots are not ready
38
128000
2000
Dupa cum vedeti acesti roboti nu sunt gata inca
02:35
to take over the world yet, but
39
130000
3000
sa preia controlul asupra lumii,dar
02:38
they gradually learn how to move forward,
40
133000
2000
invata treptat cum sa se miste inainte,
02:40
and they do this autonomously.
41
135000
3000
si fac aceste lucru in mod autonom.
02:43
So in these two examples, we had basically
42
138000
4000
Deci in aceste doua exemple, am avut de fapt
02:47
machines that learned how to walk in simulation,
43
142000
3000
masinarii care au invatat cum sa mearga intr-o simulare,
02:50
and also machines that learned how to walk in reality.
44
145000
2000
si masinarii care au invatat sa mearga in realitate.
02:52
But I want to show you a different approach,
45
147000
2000
Dar vreau sa va arat o abordare diferita,
02:54
and this is this robot over here, which has four legs.
46
149000
6000
si acesta este robotul, aici, care are patru picioare,
03:00
It has eight motors, four on the knees and four on the hip.
47
155000
2000
are opt motoare , patru la genunchi si patru la solduri.
03:02
It has also two tilt sensors that tell the machine
48
157000
3000
Mai are si doi senzori care ii spun masinariei
03:05
which way it's tilting.
49
160000
3000
in ce parte sa se incline.
03:08
But this machine doesn't know what it looks like.
50
163000
2000
Dar aceasta masinarie nu stie cum arata.
03:10
You look at it and you see it has four legs,
51
165000
2000
Tu te uiti la ea si vezi ca are patru picioare,
03:12
the machine doesn't know if it's a snake, if it's a tree,
52
167000
2000
masinaria nu stie daca e un sarpe, daca e un copac,
03:14
it doesn't have any idea what it looks like,
53
169000
3000
nu are nici o idee despre cum arata,
03:17
but it's going to try to find that out.
54
172000
2000
dar o sa incerce sa afle.
03:19
Initially, it does some random motion,
55
174000
2000
Initial, o sa incerce niste miscari aleatorii,
03:21
and then it tries to figure out what it might look like.
56
176000
3000
si apoi incearca sa afle cum arata --
03:24
And you're seeing a lot of things passing through its minds,
57
179000
2000
si vedeti cum o gramada de lucruri ii trec prin minte,
03:26
a lot of self-models that try to explain the relationship
58
181000
4000
o gramada de auto-modele care incearca sa explice relatia
03:30
between actuation and sensing. It then tries to do
59
185000
3000
dintre actiune si raspuns-- si apoi incearca
03:33
a second action that creates the most disagreement
60
188000
4000
o a doua actiune care creaza dezacord
03:37
among predictions of these alternative models,
61
192000
2000
printre predictiile modelelor alternative,
03:39
like a scientist in a lab. Then it does that
62
194000
2000
ca un om de stiinta intr-un laborator. Apoi face asta
03:41
and tries to explain that, and prune out its self-models.
63
196000
4000
si incearca sa explice, si sa isi intreaca concurentii.
03:45
This is the last cycle, and you can see it's pretty much
64
200000
3000
Acesta e ultimul ciclu, si dupa cum puteti vedea
03:48
figured out what its self looks like. And once it has a self-model,
65
203000
4000
si-a dat seama cum arata,odata ce a avut un model dupa care sa se ia,
03:52
it can use that to derive a pattern of locomotion.
66
207000
4000
se poate lua dupa asta ca sa isi creeze un tipar de locomotie.
03:56
So what you're seeing here are a couple of machines --
67
211000
2000
Deci ce vedeti aici este o adunatura de masinarii--
03:58
a pattern of locomotion.
68
213000
2000
un tipar de locomotie.
04:00
We were hoping that it wass going to have a kind of evil, spidery walk,
69
215000
4000
Speram ca o sa aiba un mers "smecher" ,ca al unui paianjen,
04:04
but instead it created this pretty lame way of moving forward.
70
219000
4000
dar in schimb,si-a creat acest mod nasol de a se misca inspre inainte.
04:08
But when you look at that, you have to remember
71
223000
3000
Dar cand te uiti la asta , trebuie sa tii cont
04:11
that this machine did not do any physical trials on how to move forward,
72
226000
6000
ca aceasta masinarie nu stia cum sa se miste inainte,
04:17
nor did it have a model of itself.
73
232000
2000
nici nu stia cum arata.
04:19
It kind of figured out what it looks like, and how to move forward,
74
234000
3000
Si-a dat seama cum arata , si cum sa se miste,
04:22
and then actually tried that out.
75
237000
4000
si apoi a facut o incercare.
04:26
(Applause)
76
241000
5000
(Aplauze)
04:31
So, we'll move forward to a different idea.
77
246000
4000
Asa, se ne indreptam atentia spre o idee diferita.
04:35
So that was what happened when we had a couple of --
78
250000
5000
Deci asta sa intamplat cand am avut o gramada de --
04:40
that's what happened when you had a couple of -- OK, OK, OK --
79
255000
4000
asta sa intamplat cand am avut o gramada de -- Ok ,Ok ,Ok--
04:44
(Laughter)
80
259000
2000
(Rasete)
04:46
-- they don't like each other. So
81
261000
2000
-- nu se plac.Deci
04:48
there's a different robot.
82
263000
3000
e un robot diferit.
04:51
That's what happened when the robots actually
83
266000
2000
Asta sa intamplat cand robotii
04:53
are rewarded for doing something.
84
268000
2000
au fost recompensati pentru ca fac ceva.
04:55
What happens if you don't reward them for anything, you just throw them in?
85
270000
3000
Dar ce se intampla cand nu ii recompensezi, doar ii arunci acolo?
04:58
So we have these cubes, like the diagram showed here.
86
273000
3000
Deci avem cuburile astea,dupa cum arata diagrama asta.
05:01
The cube can swivel, or flip on its side,
87
276000
2000
Cubul poate sa pivoteze ,sau sa sara pe o parte,
05:04
and we just throw 1,000 of these cubes into a soup --
88
279000
4000
si aruncam 1,000 de cuburi de astea intr-o supa--
05:08
this is in simulation --and don't reward them for anything,
89
283000
2000
asta intr-o simulare-- si nu ii recompensam pentru nimic.
05:10
we just let them flip. We pump energy into this
90
285000
3000
ii lasam acolo sa sara. Le dam energie
05:13
and see what happens in a couple of mutations.
91
288000
3000
si vedem ce se intampla in cateva mutatii.
05:16
So, initially nothing happens, they're just flipping around there.
92
291000
3000
Initial, nimic nu se intampla, doar sar de colo colo.
05:19
But after a very short while, you can see these blue things
93
294000
4000
Dar dupa o scurta perioada de timp,puteti vedea aceste chestii albastre
05:23
on the right there begin to take over.
94
298000
2000
din dreapta incep sa preia controlul.
05:25
They begin to self-replicate. So in absence of any reward,
95
300000
4000
Incep sa se auto-reproduca.Asa ca in absenta vreunei recompense,
05:29
the intrinsic reward is self-replication.
96
304000
3000
propria recompensa este auto-reproducerea.
05:32
And we've actually built a couple of these,
97
307000
1000
Si chiar am construit cativa din astia ,
05:33
and this is part of a larger robot made out of these cubes.
98
308000
4000
si asta e o parte dintr-un robot mai mare facut din aceste cuburi,
05:37
It's an accelerated view, where you can see the robot actually
99
312000
3000
e o filmare accelerata, in care puteti vedea cum robotul
05:40
carrying out some of its replication process.
100
315000
2000
urmeaza pasii spre procesul de replicare.
05:42
So you're feeding it with more material -- cubes in this case --
101
317000
4000
Deci o hranim cu mai mult material-- cuburi in cazul de fata--
05:46
and more energy, and it can make another robot.
102
321000
3000
si mai multa energie, si poate face un alt robot.
05:49
So of course, this is a very crude machine,
103
324000
3000
Dar desigur , aceasta este o masinarie foarte primitiva,
05:52
but we're working on a micro-scale version of these,
104
327000
2000
dar lucram la versiuni microscopice ale acestor masinarii,
05:54
and hopefully the cubes will be like a powder that you pour in.
105
329000
3000
si speram ca ,cuburile o sa fie ca o pudra pe care o adaugi.
05:57
OK, so what can we learn? These robots are of course
106
332000
5000
OK, deci ce putem invatat? Acesti roboti nu sunt desigur
06:02
not very useful in themselves, but they might teach us something
107
337000
3000
foarte folositori, dar ne pot invata cate ceva
06:05
about how we can build better robots,
108
340000
3000
despre cum putem sa contruim roboti mai buni,
06:08
and perhaps how humans, animals, create self-models and learn.
109
343000
5000
si poate cum oameni , animalele, pot crea auto-modele si invata.
06:13
And one of the things that I think is important
110
348000
2000
Si unul din lucrurile pe care il consider important
06:15
is that we have to get away from this idea
111
350000
2000
este acela ca trebuie sa ne indepartam de idea
06:17
of designing the machines manually,
112
352000
2000
de a proiecta manual masinariile,
06:19
but actually let them evolve and learn, like children,
113
354000
3000
si in schimb sa le lasam sa evolueze si sa invete,precum copiii,
06:22
and perhaps that's the way we'll get there. Thank you.
114
357000
2000
si poate in felul acesta o sa reusim . Multumesc.
06:24
(Applause)
115
359000
2000
(Aplauze)
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