TED2007
Hod Lipson: Building "self-aware" robots
哈得‧立普森建造了一個"自我感知"機器人
Filmed:
Readability: 3.1
1,460,460 views
哈得‧立普森像大家展示幾個他酷斃了的小機器人，那些機器人有學習能力，可以了解自己，甚至自行複製!
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
嗯? 所以呢? 那些機器人咧?
00:27
We've been told for 40 years already that they're coming soon.
1
2000
3000
40年來一直有人告訴我們, 機器人很快就會出現在這個世界上了。
00:30
Very soon they'll be doing everything for us.
2
5000
3000
很快地, 他們會替我們做每一件事,
00:33
They'll be cooking, cleaning, buying things, shopping, building. But they aren't here.
3
8000
5000
他們會煮飯, 打掃, 買東西, 購物血拼, 蓋房子, 但是, 他們並沒有出現。
00:38
Meanwhile, we have illegal immigrants doing all the work,
4
13000
4000
現在這當兒, 我們雇用非法移民來替我們完成所有的工作,
00:42
but we don't have any robots.
5
17000
2000
但是, 我們還是沒有機器人呀!
00:44
So what can we do about that? What can we say?
6
19000
4000
所以, 對於這件事我們可以做些什麼? 或者說些什麼呢?
00:48
So I want to give a little bit of a different perspective
7
23000
4000
所以, 我想跟你們分享一些不同的觀點,
00:52
of how we can perhaps look at these things in a little bit of a different way.
8
27000
6000
看看我們能怎樣從不同的角度看待這些事。
00:58
And this is an x-ray picture
9
33000
2000
這是一張大甲蟲和瑞士名錶的X光圖,
01:00
of a real beetle, and a Swiss watch, back from '88. You look at that --
10
35000
5000
是在1988年拍攝的, 你們看看這裡----
01:05
what was true then is certainly true today.
11
40000
2000
當年確實存在的, 現在還是存在。
01:07
We can still make the pieces. We can make the right pieces.
12
42000
3000
我們還是能做出零件, 而且是對的零件,
01:10
We can make the circuitry of the right computational power,
13
45000
3000
我們可以畫出具有運算功能的電路圖,
01:13
but we can't actually put them together to make something
14
48000
3000
但是我們卻沒有辦法把他們組合在一起然後創造出一個東西,
01:16
that will actually work and be as adaptive as these systems.
15
51000
5000
而那個東西又要能夠跟這些系統一樣運作良好又具備適應能力。
01:21
So let's try to look at it from a different perspective.
16
56000
2000
那麼讓我來試著從不一樣的角度看看,
01:23
Let's summon the best designer, the mother of all designers.
17
58000
4000
我們來召喚一個設計師--他是所有設計師的老師:
01:27
Let's see what evolution can do for us.
18
62000
3000
我們來看看演化為我們做了些什麼。
01:30
So we threw in -- we created a primordial soup
19
65000
4000
我們創造一種最原始的湯汁,
01:34
with lots of pieces of robots -- with bars, with motors, with neurons.
20
69000
4000
我們丟入很多機器人的碎片, 裡面包含了拉桿, 引擎和神經,
01:38
Put them all together, and put all this under kind of natural selection,
21
73000
4000
把他們全部放在一起, 然後讓他們面對物競天擇、
01:42
under mutation, and rewarded things for how well they can move forward.
22
77000
4000
突變,並依據他們發展的情況給予獎賞。
01:46
A very simple task, and it's interesting to see what kind of things came out of that.
23
81000
6000
這是很簡單的工作,而且觀察這個演化過程也十分有趣。
01:52
So if you look, you can see a lot of different machines
24
87000
3000
仔細一瞧,你就會發現很多不一樣的機器被創造出來了
01:55
come out of this. They all move around.
25
90000
2000
他們到處走來走去,
01:57
They all crawl in different ways, and you can see on the right,
26
92000
4000
他們都往不同的方向爬,你們可以在右邊
02:01
that we actually made a couple of these things,
27
96000
2000
看到我們做出來的成果,
02:03
and they work in reality. These are not very fantastic robots,
28
98000
3000
他們都可以在現實生活中執行任務, 你看到的這些都不是多高檔的機器人,
02:06
but they evolved to do exactly what we reward them for:
29
101000
4000
但是他們卻完全依照我們所給的獎賞而演化。
02:10
for moving forward. So that was all done in simulation,
30
105000
3000
這些雖然都是在電腦上模擬出來的,
02:13
but we can also do that on a real machine.
31
108000
2000
但我們也可以讓真的機器做出相同的事。
02:15
Here's a physical robot that we actually
32
110000
5000
這就是一個我們實際上可以看到的機器人,
02:20
have a population of brains,
33
115000
3000
他有好幾個大腦,
02:23
competing, or evolving on the machine.
34
118000
2000
這幾個大腦在這機器上彼此競爭並且演化,
02:25
It's like a rodeo show. They all get a ride on the machine,
35
120000
3000
就像賽馬一樣: 他們會騎到機器上,
02:28
and they get rewarded for how fast or how far
36
123000
3000
他們之中讓機器跑得越快越遠的,
02:31
they can make the machine move forward.
37
126000
2000
就可以得到越多獎賞。
02:33
And you can see these robots are not ready
38
128000
2000
現在你可以看到這些機器人
02:35
to take over the world yet, but
39
130000
3000
還沒準備好取代人類統治這個世界,
02:38
they gradually learn how to move forward,
40
133000
2000
不過他們慢慢學會要怎麼往前走了,
02:40
and they do this autonomously.
41
135000
3000
而且他們是完全自主地前進著。
02:43
So in these two examples, we had basically
42
138000
4000
在剛剛提到的兩個例子裡, 我們基本上擁有兩種機器,
02:47
machines that learned how to walk in simulation,
43
142000
3000
第一種是以電腦模擬的方式學習走路,
02:50
and also machines that learned how to walk in reality.
44
145000
2000
第二種則是在現實生活中學習前進。
02:52
But I want to show you a different approach,
45
147000
2000
但是我要讓你們看的是另一種更不一樣的方式,
02:54
and this is this robot over here, which has four legs.
46
149000
6000
請看, 這隻機器人有四隻腳,
03:00
It has eight motors, four on the knees and four on the hip.
47
155000
2000
八個引擎, 四個在膝蓋的地方, 另外四個在臀部。
03:02
It has also two tilt sensors that tell the machine
48
157000
3000
他還有兩個傾斜感應器,
03:05
which way it's tilting.
49
160000
3000
用來感應自己向哪裡傾斜了。
03:08
But this machine doesn't know what it looks like.
50
163000
2000
但是這機器並不知道他自己長什麼樣子,
03:10
You look at it and you see it has four legs,
51
165000
2000
你看得到他, 所以知道他有四隻腳。
03:12
the machine doesn't know if it's a snake, if it's a tree,
52
167000
2000
但是這機器卻沒辦法知道他自己是一條蛇還是一顆樹,
03:14
it doesn't have any idea what it looks like,
53
169000
3000
他完全不清楚自己長什麼樣子,
03:17
but it's going to try to find that out.
54
172000
2000
但是他會想辦法知道。
03:19
Initially, it does some random motion,
55
174000
2000
一開始, 他會隨機做一些動作,
03:21
and then it tries to figure out what it might look like.
56
176000
3000
接著他試圖看出自己大概長什麼樣子--
03:24
And you're seeing a lot of things passing through its minds,
57
179000
2000
你會看到他腦海中浮現非常多東西,
03:26
a lot of self-models that try to explain the relationship
58
181000
4000
有很多他自己創造的動作模式, 他試圖去釐清
03:30
between actuation and sensing. It then tries to do
59
185000
3000
動作和感知之間的關係─然後他再試著做第二個動作,
03:33
a second action that creates the most disagreement
60
188000
4000
那個動作不在既有的動作模式內,
03:37
among predictions of these alternative models,
61
192000
2000
完全出乎我們的意料,
03:39
like a scientist in a lab. Then it does that
62
194000
2000
就像在實驗室裡的科學家一樣。接著, 他重複那個動作,
03:41
and tries to explain that, and prune out its self-models.
63
196000
4000
並且試著解釋那個動作, 然後創造出自己的動作模式。
03:45
This is the last cycle, and you can see it's pretty much
64
200000
3000
這是他最後一次重覆這整個循環, 你可以看到他
03:48
figured out what its self looks like. And once it has a self-model,
65
203000
4000
已經弄清楚自己的樣子了, 一旦他整理出自己的動作模式,
03:52
it can use that to derive a pattern of locomotion.
66
207000
4000
就可以從中發展出一種運動模式。
03:56
So what you're seeing here are a couple of machines --
67
211000
2000
你現在看到的是幾個機器─
03:58
a pattern of locomotion.
68
213000
2000
一種運動模式。
04:00
We were hoping that it wass going to have a kind of evil, spidery walk,
69
215000
4000
我們期待他做出一種如惡魔或者蜘蛛般的行走模式,
04:04
but instead it created this pretty lame way of moving forward.
70
219000
4000
但是他卻創造出這種看似殘障的前進方法。
04:08
But when you look at that, you have to remember
71
223000
3000
但是當你看著他前進的時候, 你必須記得,
04:11
that this machine did not do any physical trials on how to move forward,
72
226000
6000
這機器並沒有做過任何往前行進的物理試驗,
04:17
nor did it have a model of itself.
73
232000
2000
他也沒有任何屬於自己的模式,
04:19
It kind of figured out what it looks like, and how to move forward,
74
234000
3000
他等於是自己發現了自己的樣子, 然後找出前進的方法,
04:22
and then actually tried that out.
75
237000
4000
並實際驗證成功。
04:26
(Applause)
76
241000
5000
(掌聲響起)
04:31
So, we'll move forward to a different idea.
77
246000
4000
那麼現在, 我們再來看看另一種想法,
04:35
So that was what happened when we had a couple of --
78
250000
5000
那是我們將幾個─
04:40
that's what happened when you had a couple of -- OK, OK, OK --
79
255000
4000
把幾個放在一塊兒就會......好啦好啦好啦!!!
04:44
(Laughter)
80
259000
2000
(笑聲)
04:46
-- they don't like each other. So
81
261000
2000
─他們不太喜歡對方, 所以
04:48
there's a different robot.
82
263000
3000
這是另一個機器人。
04:51
That's what happened when the robots actually
83
266000
2000
剛剛的事情都是因為機器人做對動作,
04:53
are rewarded for doing something.
84
268000
2000
並且得到獎勵才發生的。
04:55
What happens if you don't reward them for anything, you just throw them in?
85
270000
3000
那如果們不給他們獎勵, 直接把他們丟在一塊會怎麼樣呢?
04:58
So we have these cubes, like the diagram showed here.
86
273000
3000
所以我們拿來了這些立方體, 就像你在圖上看到的,
05:01
The cube can swivel, or flip on its side,
87
276000
2000
他們會旋轉或者翻轉。
05:04
and we just throw 1,000 of these cubes into a soup --
88
279000
4000
我們把一千個這樣的立方體放入"原始湯汁"裡─
05:08
this is in simulation --and don't reward them for anything,
89
283000
2000
這是電腦模擬效果─我們沒有給他們任何獎勵,
05:10
we just let them flip. We pump energy into this
90
285000
3000
就讓他們翻轉而已。我們給他們一些能量,
05:13
and see what happens in a couple of mutations.
91
288000
3000
看看經過幾次突變以後會怎樣。
05:16
So, initially nothing happens, they're just flipping around there.
92
291000
3000
剛開始什麼都沒發生, 他們就只是跳來跳去,
05:19
But after a very short while, you can see these blue things
93
294000
4000
但又過了一下, 你就會看到右邊那些
05:23
on the right there begin to take over.
94
298000
2000
藍色的東西開始掌控全局。
05:25
They begin to self-replicate. So in absence of any reward,
95
300000
4000
他們開始自我複製, 由此可見就算沒有獎勵,
05:29
the intrinsic reward is self-replication.
96
304000
3000
他們也會用自我複製的方式獎勵自己。
05:32
And we've actually built a couple of these,
97
307000
1000
實際上我們已經製造了好幾個像這樣的玩意兒,
05:33
and this is part of a larger robot made out of these cubes.
98
308000
4000
這是用這樣的立方體做出來的機器人其中的一部分,
05:37
It's an accelerated view, where you can see the robot actually
99
312000
3000
我們用快轉的方式,讓你看看這機器人
05:40
carrying out some of its replication process.
100
315000
2000
進行自我複製的過程。
05:42
So you're feeding it with more material -- cubes in this case --
101
317000
4000
如果你多餵給這個機器人一些原料─那些立方體─
05:46
and more energy, and it can make another robot.
102
321000
3000
還有很多能量, 他可以製造出另一個機器人。
05:49
So of course, this is a very crude machine,
103
324000
3000
當然, 這是一個很粗糙的機器,
05:52
but we're working on a micro-scale version of these,
104
327000
2000
但是我們正努力做出這種機器人的縮小版,
05:54
and hopefully the cubes will be like a powder that you pour in.
105
329000
3000
希望這些立方體可以小到跟粉末一樣。
05:57
OK, so what can we learn? These robots are of course
106
332000
5000
好吧!那麼我們學到了些什麼呢?這些機器人
06:02
not very useful in themselves, but they might teach us something
107
337000
3000
本身不一定多有用, 但他們卻可以教會我們一些事情,
06:05
about how we can build better robots,
108
340000
3000
關於我們如何做出更好的機器人,
06:08
and perhaps how humans, animals, create self-models and learn.
109
343000
5000
甚至是人類或動物創造自我模式跟學習的機制原理。
06:13
And one of the things that I think is important
110
348000
2000
還有一樣我覺得最重要的，
06:15
is that we have to get away from this idea
111
350000
2000
就是我們要放棄
06:17
of designing the machines manually,
112
352000
2000
以人工設計機器的想法，
06:19
but actually let them evolve and learn, like children,
113
354000
3000
放手讓機器自己演化與學習, 像孩子一樣，
06:22
and perhaps that's the way we'll get there. Thank you.
114
357000
2000
這或許才是讓我們成功的辦法, 謝謝！
06:24
(Applause)
115
359000
2000
(掌聲)
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