TED2007
Hod Lipson: Building "self-aware" robots
霍德·利普森建造的“自我感知”机器人
Filmed:
Readability: 3.1
1,460,460 views
霍德·利普森演示了他的几个神奇的小机器人。它们不仅有学习能力，理解自己，甚至还能进行自我复制！
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
啊，机器人们都在哪儿呢？
00:27
We've been told for 40 years already that they're coming soon.
1
2000
3000
四十年前人们就说它们很快就要来了
00:30
Very soon they'll be doing everything for us.
2
5000
3000
很快，它们就能为我们做一切事情
00:33
They'll be cooking, cleaning, buying things, shopping, building. But they aren't here.
3
8000
5000
它们会做饭，打扫，买东西，购物，甚至是建房子。但直到今天，它们也没能进入我们的生活。
00:38
Meanwhile, we have illegal immigrants doing all the work,
4
13000
4000
这段时间里，非法移民承担着这些工作，
00:42
but we don't have any robots.
5
17000
2000
但我们什么机器人都没有。
00:44
So what can we do about that? What can we say?
6
19000
4000
我们又能做些或说些什么呢？
00:48
So I want to give a little bit of a different perspective
7
23000
4000
我想给大家带来点不同的启发
00:52
of how we can perhaps look at these things in a little bit of a different way.
8
27000
6000
看看今天我们能不能换一个角度看待这些事情
00:58
And this is an x-ray picture
9
33000
2000
这是一张X光片
01:00
of a real beetle, and a Swiss watch, back from '88. You look at that --
10
35000
5000
上面有一只活甲虫，和一只88年的瑞士手表，你看——
01:05
what was true then is certainly true today.
11
40000
2000
无论是当时还是现在
01:07
We can still make the pieces. We can make the right pieces.
12
42000
3000
我们都可以做出这些零件，一模一样的零件，
01:10
We can make the circuitry of the right computational power,
13
45000
3000
我们能做出有同样计算能力的电路，
01:13
but we can't actually put them together to make something
14
48000
3000
但我们并不能把它们放在一起，再做出个什么东西
01:16
that will actually work and be as adaptive as these systems.
15
51000
5000
能和这些系统（甲虫）有一样的适应力。
01:21
So let's try to look at it from a different perspective.
16
56000
2000
那么就让我们试着换个角度再看看这个问题。
01:23
Let's summon the best designer, the mother of all designers.
17
58000
4000
让我们召集最好的设计师，所有设计师的鼻祖，
01:27
Let's see what evolution can do for us.
18
62000
3000
看看进化论能给我们做些什么。
01:30
So we threw in -- we created a primordial soup
19
65000
4000
所以我们就找来了--我们创造了一锅“原汤”，
01:34
with lots of pieces of robots -- with bars, with motors, with neurons.
20
69000
4000
里面有组装机器人需要的各种零件：条状的，带马达的，还有些带神经元的。
01:38
Put them all together, and put all this under kind of natural selection,
21
73000
4000
把它们都放在一起，然后把所有这些放到一种自然选择，
01:42
under mutation, and rewarded things for how well they can move forward.
22
77000
4000
可能产生突变的环境中，并奖励那些发生“进化”的零件组。
01:46
A very simple task, and it's interesting to see what kind of things came out of that.
23
81000
6000
这是一个非常简单的任务，并且得出的结果非常有趣。
01:52
So if you look, you can see a lot of different machines
24
87000
3000
看一下，你能看到各式各样的机器
01:55
come out of this. They all move around.
25
90000
2000
在这种模式下制造出来，它们到处移动，
01:57
They all crawl in different ways, and you can see on the right,
26
92000
4000
以不同的方式爬行，在右边你可以看到
02:01
that we actually made a couple of these things,
27
96000
2000
我们真的做出了几个这样的玩意儿，
02:03
and they work in reality. These are not very fantastic robots,
28
98000
3000
它们在现实中真的能工作。这些还算不上非常先进的机器人，
02:06
but they evolved to do exactly what we reward them for:
29
101000
4000
但它们的确按照我们奖励的方向进化了：
02:10
for moving forward. So that was all done in simulation,
30
105000
3000
那就是向前进化。这些都是模拟的，
02:13
but we can also do that on a real machine.
31
108000
2000
但我们在真的机器上也做成功了，
02:15
Here's a physical robot that we actually
32
110000
5000
这是一个真实的机器人，
02:20
have a population of brains,
33
115000
3000
我们动用了大量的人力脑力
02:23
competing, or evolving on the machine.
34
118000
2000
让机器们相互竞争，共同进化
02:25
It's like a rodeo show. They all get a ride on the machine,
35
120000
3000
这有点像一场套马表演：他们都骑在机器上
02:28
and they get rewarded for how fast or how far
36
123000
3000
根据他们能够驾驶机器前进的速度和距离
02:31
they can make the machine move forward.
37
126000
2000
得到奖励
02:33
And you can see these robots are not ready
38
128000
2000
你可以看到这些机器人都还没完全准备好
02:35
to take over the world yet, but
39
130000
3000
占领这个世界，但是
02:38
they gradually learn how to move forward,
40
133000
2000
它们渐渐地学会如何前进
02:40
and they do this autonomously.
41
135000
3000
并且是自发地学习。
02:43
So in these two examples, we had basically
42
138000
4000
所以从这两个例子中，我们已经基本上
02:47
machines that learned how to walk in simulation,
43
142000
3000
得到了能在虚拟中学习走路
02:50
and also machines that learned how to walk in reality.
44
145000
2000
和在现实中学习走路的机器人。
02:52
But I want to show you a different approach,
45
147000
2000
现在我还要再给你们展示另一项进展，
02:54
and this is this robot over here, which has four legs.
46
149000
6000
就是这里的这个机器人，在这儿，它有四条腿
03:00
It has eight motors, four on the knees and four on the hip.
47
155000
2000
它身上安了八个马达，膝盖上四个，腿上四个
03:02
It has also two tilt sensors that tell the machine
48
157000
3000
它还配备了两个倾斜度传感器，可以告诉自己
03:05
which way it's tilting.
49
160000
3000
正在向哪个方向倾斜。
03:08
But this machine doesn't know what it looks like.
50
163000
2000
但这个机器并不知道自己长啥样
03:10
You look at it and you see it has four legs,
51
165000
2000
你可以看到它长了四条腿，
03:12
the machine doesn't know if it's a snake, if it's a tree,
52
167000
2000
但它自己并不知道自己是一条蛇还是一棵树
03:14
it doesn't have any idea what it looks like,
53
169000
3000
它完全给蒙在鼓里，不晓得自己的相貌
03:17
but it's going to try to find that out.
54
172000
2000
但它马上就要试着找到自己
03:19
Initially, it does some random motion,
55
174000
2000
首先，它会做一些随机的动作
03:21
and then it tries to figure out what it might look like.
56
176000
3000
然后试着弄清楚那些动作都看起来是什么样子的——
03:24
And you're seeing a lot of things passing through its minds,
57
179000
2000
你可以看到它的脑海里闪过许许多多的东西，
03:26
a lot of self-models that try to explain the relationship
58
181000
4000
大量的自我尝试的动作模型，试着理清
03:30
between actuation and sensing. It then tries to do
59
185000
3000
行动和感官之间的关系——然后它将再做第二个动作
03:33
a second action that creates the most disagreement
60
188000
4000
在所有可能的动作模型中
03:37
among predictions of these alternative models,
61
192000
2000
最诡异的一个动作，
03:39
like a scientist in a lab. Then it does that
62
194000
2000
就像科学家在实验室里的试验。接着，它重复那个动作，
03:41
and tries to explain that, and prune out its self-models.
63
196000
4000
并且试着解释那个动作，然后梳理出自己的动作模型。
03:45
This is the last cycle, and you can see it's pretty much
64
200000
3000
这是最后一个环节，你可以看到它已经基本上
03:48
figured out what its self looks like. And once it has a self-model,
65
203000
4000
清楚自己的样子了。一旦它理清自己的动作模型，
03:52
it can use that to derive a pattern of locomotion.
66
207000
4000
就可以从模型中得出一种运动模式。
03:56
So what you're seeing here are a couple of machines --
67
211000
2000
好了，你现在看到的是几个机器——
03:58
a pattern of locomotion.
68
213000
2000
嗯，一种运动模式，
04:00
We were hoping that it wass going to have a kind of evil, spidery walk,
69
215000
4000
我们期待它能产生一种邪恶的，蜘蛛式的运动，
04:04
but instead it created this pretty lame way of moving forward.
70
219000
4000
结果它却自创出这种相当脑残的前进方式。
04:08
But when you look at that, you have to remember
71
223000
3000
但当你看着它前进的时候，你必须记得
04:11
that this machine did not do any physical trials on how to move forward,
72
226000
6000
这个机器并没有接受任何物理指令，控制着它们向前进，
04:17
nor did it have a model of itself.
73
232000
2000
它也没有任何已有的自我模型。
04:19
It kind of figured out what it looks like, and how to move forward,
74
234000
3000
它相当于是自己推理出了自己的样子，以及应该如何向前进
04:22
and then actually tried that out.
75
237000
4000
并且进行了亲身的尝试。
04:26
(Applause)
76
241000
5000
鼓掌~~~♫
04:31
So, we'll move forward to a different idea.
77
246000
4000
那么现在，我们再来看看另一个想法。
04:35
So that was what happened when we had a couple of --
78
250000
5000
那是我们将几个...
04:40
that's what happened when you had a couple of -- OK, OK, OK --
79
255000
4000
把几个...放在一块儿就会...好啦好啦好啦——
04:44
(Laughter)
80
259000
2000
（笑）
04:46
-- they don't like each other. So
81
261000
2000
——它们不大喜欢对方，所以啦~
04:48
there's a different robot.
82
263000
3000
这是另外一个机器人。
04:51
That's what happened when the robots actually
83
266000
2000
刚才的那些都是在机器人做对了动作，
04:53
are rewarded for doing something.
84
268000
2000
获得奖励的情况下发生的。
04:55
What happens if you don't reward them for anything, you just throw them in?
85
270000
3000
那么如果我们不给它们奖励，只是把它们扔到一块，又会怎么样呢？
04:58
So we have these cubes, like the diagram showed here.
86
273000
3000
所以我们拿来了这些立方体，就像这里的这些图，
05:01
The cube can swivel, or flip on its side,
87
276000
2000
它们能旋转，或者翻筋斗
05:04
and we just throw 1,000 of these cubes into a soup --
88
279000
4000
我们把1000个这样的立方体放入“原汤”——
05:08
this is in simulation --and don't reward them for anything,
89
283000
2000
这是模拟效果——我们没给它们任何奖励，
05:10
we just let them flip. We pump energy into this
90
285000
3000
我们就让它们自己活动。我们给它们注入了些能量，
05:13
and see what happens in a couple of mutations.
91
288000
3000
看看经过几次突变，会发生点什么。
05:16
So, initially nothing happens, they're just flipping around there.
92
291000
3000
刚开始的时候，什么也没发生，它们光在那儿跳来跳去。
05:19
But after a very short while, you can see these blue things
93
294000
4000
但过了一小会儿，你可以看到这些蓝色的小东西，
05:23
on the right there begin to take over.
94
298000
2000
它们在右边逐渐地开始占取主动。
05:25
They begin to self-replicate. So in absence of any reward,
95
300000
4000
它们开始自我复制。由此可见即使没有任何奖励
05:29
the intrinsic reward is self-replication.
96
304000
3000
它们也会用自我复制的方式来奖励自己。
05:32
And we've actually built a couple of these,
97
307000
1000
事实上我们已经造了好几个这类的玩意儿，
05:33
and this is part of a larger robot made out of these cubes.
98
308000
4000
这是一部分以这些立方体为单位造出来的大机器人，
05:37
It's an accelerated view, where you can see the robot actually
99
312000
3000
这是快进的效果，可以让你看到机器人
05:40
carrying out some of its replication process.
100
315000
2000
进行自我复制的过程。
05:42
So you're feeding it with more material -- cubes in this case --
101
317000
4000
如果你给它多喂点儿——就是这些立方体——
05:46
and more energy, and it can make another robot.
102
321000
3000
再多给它点能量，它就能自己造出另一个机器人。
05:49
So of course, this is a very crude machine,
103
324000
3000
当然，这还是一个非常粗糙，不成熟的机器，
05:52
but we're working on a micro-scale version of these,
104
327000
2000
但我们正在研究微缩版的这类机器人，
05:54
and hopefully the cubes will be like a powder that you pour in.
105
329000
3000
希望这些立方体能小到像倒出的面粉一般。
05:57
OK, so what can we learn? These robots are of course
106
332000
5000
好的，那么我们都了解到了什么？这些机器人当然
06:02
not very useful in themselves, but they might teach us something
107
337000
3000
自己并不是有很大用处，但它们能教会我们一些东西，
06:05
about how we can build better robots,
108
340000
3000
让我们知道如何造出更好的机器人，
06:08
and perhaps how humans, animals, create self-models and learn.
109
343000
5000
甚至是人类，动物创造自我模型和学习机制的原理。
06:13
And one of the things that I think is important
110
348000
2000
我觉得这其中最重要的，
06:15
is that we have to get away from this idea
111
350000
2000
就是我们必须摒弃之前的观念，
06:17
of designing the machines manually,
112
352000
2000
手动地设计这些机器
06:19
but actually let them evolve and learn, like children,
113
354000
3000
而是让它们自己进化，学习，像孩子一样，
06:22
and perhaps that's the way we'll get there. Thank you.
114
357000
2000
这大概才是我们成功的必经之路。谢谢！
06:24
(Applause)
115
359000
2000
（鼓掌♫）
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