TEDxNASA
Dennis Hong: My seven species of robot -- and how we created them
Dennis Hong：我的七種機器人
Filmed:
Readability: 4.2
2,237,651 views
在 TEDxNASA，Dennis Hong 介紹了七個獲獎的全地形機器人，例如：人形化的足球機器人 DARwin 和攀岩機器人 CLIMBeR。全都由他的團隊在維吉尼亞理工的 RoMeLa 實驗室所研製出來。請耐心看下去，聆聽那五個讓他的實驗室取得如此驚人成功的創意秘訣。
Dennis Hong - Roboticist
Dennis Hong is the founder and director of RoMeLa -- a Virginia Tech robotics lab that has pioneered several breakthroughs in robot design and engineering. Full bio
Dennis Hong is the founder and director of RoMeLa -- a Virginia Tech robotics lab that has pioneered several breakthroughs in robot design and engineering. Full bio
Double-click the English transcript below to play the video.
00:15
So, the first robot to talk about is called STriDER.
0
0
3000
我們首先談到的機器人叫 STriDER，
00:18
It stands for Self-excited
1
3000
2000
它的意思是
00:20
Tripedal Dynamic Experimental Robot.
2
5000
2000
三腳動力實驗性機器人。
00:22
It's a robot that has three legs,
3
7000
2000
這是一個有三隻腳的機器人，
00:24
which is inspired by nature.
4
9000
3000
它的靈感是從自然界中所獲得的。
00:27
But have you seen anything in nature,
5
12000
2000
但是，你可曾看過自然界中
00:29
an animal that has three legs?
6
14000
2000
有三隻腳的生物嗎？
00:31
Probably not. So, why do I call this
7
16000
2000
大概沒有，那爲什麽我稱這機器人為
00:33
a biologically inspired robot? How would it work?
8
18000
2000
被生物所啓發的機器人？這機器人是怎麼運作的呢？
00:35
But before that, let's look at pop culture.
9
20000
3000
在我解釋之前，讓我們先來看看流行文化，
00:38
So, you know H.G. Wells' "War of the Worlds," novel and movie.
10
23000
3000
都聽過 H.G.Wells 的世界大戰小說和電影吧。
00:41
And what you see over here is a very popular
11
26000
2000
你們現在在這邊所看到的是一個很熱門的
00:43
video game,
12
28000
2000
電腦遊戲。
00:45
and in this fiction they describe these alien creatures that
13
30000
3000
在小說裏面，他們把這些襲擊地球的
00:48
are robots that have three legs that terrorize Earth.
14
33000
2000
外星生物描述成有三隻腳的機器人。
00:50
But my robot, STriDER, does not move like this.
15
35000
4000
但是我的機器人 STriDER，不是這樣移動的。
00:54
So, this is an actual dynamic simulation animation.
16
39000
3000
這是一個真正的動力學模擬動畫。
00:57
I'm just going to show you how the robot works.
17
42000
2000
我將會向你們展示這個機器人是如何運作的。
00:59
It flips its body 180 degrees
18
44000
3000
它把自己翻了180度，
01:02
and it swings its leg between the two legs and catches the fall.
19
47000
3000
它通過甩動兩條腿中間的腿來防止自己跌倒。
01:05
So, that's how it walks. But when you look at us
20
50000
2000
看，它就是這樣走路的。但是，我們再看看自己，
01:07
human being, bipedal walking,
21
52000
2000
人類 — 靠雙腿行走的生物，
01:09
what you're doing is you're not really using a muscle
22
54000
2000
我們並沒有用肌肉
01:11
to lift your leg and walk like a robot. Right?
23
56000
3000
抬起自己的腿，然後走得像個機器人，對吧？
01:14
What you're doing is you really swing your leg and catch the fall,
24
59000
3000
你做的實際上是甩開你的腿來止住自己跌倒的趨勢，
01:17
stand up again, swing your leg and catch the fall.
25
62000
3000
重新站穩，并再次甩開腿。
01:20
You're using your built-in dynamics, the physics of your body,
26
65000
3000
通過運用我們身體內置的動力機制 — 身體的物理構造，
01:23
just like a pendulum.
27
68000
2000
就像個鐘擺一般，
01:25
We call that the concept of passive dynamic locomotion.
28
70000
4000
我們把這概念叫做消極動力運動。
01:29
What you're doing is, when you stand up,
29
74000
2000
你做的是，當你站立的時候，
01:31
potential energy to kinetic energy,
30
76000
2000
位能轉化
01:33
potential energy to kinetic energy.
31
78000
2000
為動能，
01:35
It's a constantly falling process.
32
80000
2000
這是一個普通的下落過程。
01:37
So, even though there is nothing in nature that looks like this,
33
82000
3000
所以，儘管自然界中沒有生物長成這樣，
01:40
really, we were inspired by biology
34
85000
2000
我們仍然從生物學中受到啓發，
01:42
and applying the principles of walking
35
87000
2000
并製造出機器人的這種行走模式。
01:44
to this robot. Thus it's a biologically inspired robot.
36
89000
3000
所以，這是一個受生物所啓發的機器人。
01:47
What you see over here, this is what we want to do next.
37
92000
2000
你們現在所看到的這個，則是我們接下來想要做的。
01:49
We want to fold up the legs and shoot it up for long-range motion.
38
94000
4000
我們想把腿折起來，然後把它射出去以做長距離的移動。
01:53
And it deploys legs -- it looks almost like "Star Wars" --
39
98000
3000
它叉開它的腿，這讓它看起來像電影星際大戰裡面的角色。
01:56
when it lands, it absorbs the shock and starts walking.
40
101000
3000
當它降落的時候，它能吸收震盪，并開始行走。
01:59
What you see over here, this yellow thing, this is not a death ray. (Laughter)
41
104000
3000
你們在這邊所看到的這個，這黃色的東西，這不是死光。
02:02
This is just to show you that if you have cameras
42
107000
2000
這只是用來告訴你，機器人上是否具備相機，
02:04
or different types of sensors --
43
109000
2000
或別種感測器，
02:06
because it is tall, it's 1.8 meters tall --
44
111000
2000
因為它很高，它有 1.8 公尺高，
02:08
you can see over obstacles like bushes and those kinds of things.
45
113000
3000
所以你可以穿過像灌木叢這樣的障礙物來觀察前方。
02:11
So we have two prototypes.
46
116000
2000
我們做了兩個原型機，
02:13
The first version, in the back, that's STriDER I.
47
118000
3000
第一版，後面的那個，這是 STriDER I，
02:16
The one in front, the smaller, is STriDER II.
48
121000
2000
這在前面的，更小的，是 STriDER II.
02:18
The problem that we had with STriDER I is
49
123000
2000
STriDER I 的問題是
02:20
it was just too heavy in the body. We had so many motors,
50
125000
3000
身體過重。因為你知道，我們要
02:23
you know, aligning the joints, and those kinds of things.
51
128000
2000
連接許多關節，所以安裝上很多馬達一類的東西。
02:25
So, we decided to synthesize a mechanical mechanism
52
130000
4000
所以，我們決定想出一種機械原理，
02:29
so we could get rid of all the motors, and with a single motor
53
134000
3000
以擺脫所有的馬達，只需使用單一馬達，
02:32
we can coordinate all the motions.
54
137000
2000
我們就可以連貫所有動作。
02:34
It's a mechanical solution to a problem, instead of using mechatronics.
55
139000
3000
我們運用了機械原理來解決問題，而不是使用電子機械。
02:37
So, with this now the top body is light enough. So, it's walking in our lab;
56
142000
3000
現在，機器人的上身輕得足夠讓它在實驗室行走。
02:40
this was the very first successful step.
57
145000
3000
這只是邁向成功的第一步。
02:43
It's still not perfected -- its coffee falls down --
58
148000
2000
它仍然是不完美的。它拿著的咖啡掉了，
02:45
so we still have a lot of work to do.
59
150000
3000
所以，我們仍然有很多工作要做。
02:48
The second robot I want to talk about is called IMPASS.
60
153000
3000
我要介紹的第二個機器人叫 IMPASS。
02:51
It stands for Intelligent Mobility Platform with Actuated Spoke System.
61
156000
4000
它的意思是輪輻驅動系統的智慧型移動平臺，
02:55
So, it's a wheel-leg hybrid robot.
62
160000
3000
這是個輪、腳混合的機器人。
02:58
So, think of a rimless wheel
63
163000
2000
試著想像一個沒輪框的輪子，
03:00
or a spoke wheel,
64
165000
2000
或者一個有輻條的輪子，
03:02
but the spokes individually move in and out of the hub;
65
167000
3000
但是輪輻獨立地在輪軸裡外運動，
03:05
so, it's a wheel-leg hybrid.
66
170000
2000
所以就成了輪、腳混合，
03:07
We are literally re-inventing the wheel here.
67
172000
2000
事實上，我們相當於重新發明了一種輪子。
03:09
Let me demonstrate how it works.
68
174000
3000
讓我來示範一下它是如何運作的。
03:12
So, in this video we're using an approach
69
177000
2000
在這個影片裡，我們用了一種方法，
03:14
called the reactive approach.
70
179000
2000
叫反應途徑。
03:16
Just simply using the tactile sensors on the feet,
71
181000
3000
只要運用它腳上的觸覺感應器，
03:19
it's trying to walk over a changing terrain,
72
184000
2000
它正嘗試穿過一個不停改變的地形，
03:21
a soft terrain where it pushes down and changes.
73
186000
3000
一個柔軟的地形，它會受壓變形。
03:24
And just by the tactile information,
74
189000
2000
僅憑藉它的觸覺訊息，
03:26
it successfully crosses over these type of terrain.
75
191000
3000
它成功地穿越了這些地形。
03:29
But, when it encounters a very extreme terrain,
76
194000
4000
但，當它遭遇了一個十分嚴峻的地形，比如說：
03:33
in this case, this obstacle is more than three times
77
198000
3000
在它前面的障礙是
03:36
the height of the robot,
78
201000
2000
它身高的3倍多，
03:38
Then it switches to a deliberate mode,
79
203000
2000
它就會轉到一種思考模式，
03:40
where it uses a laser range finder,
80
205000
2000
這個模式使用一種鐳射找尋器，
03:42
and camera systems, to identify the obstacle and the size,
81
207000
2000
以及相機系統，來確定障礙和它的體積，
03:44
and it plans, carefully plans the motion of the spokes
82
209000
3000
然後它會做計劃，小心的計劃它輪輻的移動，
03:47
and coordinates it so that it can show this
83
212000
2000
並協調其移動，因此，它就能夠做出這些
03:49
kind of very very impressive mobility.
84
214000
2000
十分讓人驚歎的移動能力。
03:51
You probably haven't seen anything like this out there.
85
216000
2000
你可能在外頭從沒看過類似這樣的東西。
03:53
This is a very high mobility robot
86
218000
3000
這是我們所開發的一種有
03:56
that we developed called IMPASS.
87
221000
3000
很高移動能力的機器人，叫做 IMPASS。
03:59
Ah, isn't that cool?
88
224000
2000
啊...這不是很酷嗎？
04:01
When you drive your car,
89
226000
3000
當你開車的時候，
04:04
when you steer your car, you use a method
90
229000
2000
當你駕駛你的汽車的時候，你用一種
04:06
called Ackermann steering.
91
231000
2000
叫做 Ackermann 的方法，
04:08
The front wheels rotate like this.
92
233000
2000
來做前輪驅動，像這樣。
04:10
For most small wheeled robots,
93
235000
3000
對於大多數的小輪子機器人，
04:13
they use a method called differential steering
94
238000
2000
它們用一種叫差速齒輪方法，
04:15
where the left and right wheel turns the opposite direction.
95
240000
3000
即左右兩個輪分別轉向相反方向。
04:18
For IMPASS, we can do many, many different types of motion.
96
243000
3000
對於 IMPASS 來說，我們可以做很多不同類型的動作。
04:21
For example, in this case, even though the left and right wheel is connected
97
246000
3000
舉例來說，在這裡，儘管左輪和右輪通過
04:24
with a single axle rotating at the same angle of velocity.
98
249000
2000
單一輪軸相連，以相同的角速度在旋轉，
04:26
We just simply change the length of the spoke.
99
251000
3000
我們只要簡單的改變輪輻的長度，
04:29
It affects the diameter and then can turn to the left, turn to the right.
100
254000
2000
它影響了直徑，然後它就會向左轉，再向右轉。
04:31
So, these are just some examples of the neat things
101
256000
2000
所以，這只是一些 IMPASS
04:33
that we can do with IMPASS.
102
258000
3000
可以做到的美妙的例子。
04:36
This robot is called CLIMBeR:
103
261000
2000
這個機器人叫 CLIMBeR，
04:38
Cable-suspended Limbed Intelligent Matching Behavior Robot.
104
263000
3000
吊索智能行為協調機器人。
04:41
So, I've been talking to a lot of NASA JPL scientists --
105
266000
3000
我曾和很多 NASA JPL 的科學家交談過，
04:44
at JPL they are famous for the Mars rovers --
106
269000
2000
他們是 JPL 登月機器人的專家。
04:46
and the scientists, geologists always tell me
107
271000
2000
這些科學家，地理學家總是告訴我，
04:48
that the real interesting science,
108
273000
3000
真正有趣的科學研究，
04:51
the science-rich sites, are always at the cliffs.
109
276000
3000
富有研究價值的地點，總是在懸崖邊緣才能發現。
04:54
But the current rovers cannot get there.
110
279000
2000
但是，現在的登月機器人到不了那些地方。
04:56
So, inspired by that we wanted to build a robot
111
281000
2000
因此，受到這對話的啟發，我們想建一個機器人，
04:58
that can climb a structured cliff environment.
112
283000
3000
能夠攀爬懸崖峭壁，
05:01
So, this is CLIMBeR.
113
286000
2000
於是，我們有了 CLIMBeR。
05:03
So, what it does, it has three legs. It's probably difficult to see,
114
288000
2000
來看看它是如何做到的，它有三條腿，這大概很難看到，
05:05
but it has a winch and a cable at the top --
115
290000
3000
它有個絞盤和繩索在頂部，
05:08
and it tries to figure out the best place to put its foot.
116
293000
2000
它會嘗試找到最好的地方落腳。
05:10
And then once it figures that out
117
295000
2000
一旦找到了這地方，
05:12
in real time, it calculates the force distribution:
118
297000
3000
它即時計算力量的分佈。
05:15
how much force it needs to exert to the surface
119
300000
3000
它需要多大的力氣才能在表面站穩，
05:18
so it doesn't tip and doesn't slip.
120
303000
2000
而不至於會摔倒或者滑到。
05:20
Once it stabilizes that, it lifts a foot,
121
305000
2000
一旦它穩定後，就會抬起一條腿，
05:22
and then with the winch it can climb up these kinds of thing.
122
307000
4000
然後，使用絞盤，它可以爬上這種東西。
05:26
Also for search and rescue applications as well.
123
311000
2000
這種技術用於搜索和救難同樣有用。
05:28
Five years ago I actually worked at NASA JPL
124
313000
2000
五年前的一個夏天我在 NASA JPL
05:30
during the summer as a faculty fellow.
125
315000
2000
兼任暑期教職。
05:32
And they already had a six legged robot called LEMUR.
126
317000
4000
他們已經有一個六支腳的機器人，叫 LEMUR。
05:36
So, this is actually based on that. This robot is called MARS:
127
321000
3000
這機器人其實是以 LEMUR 為基礎的，它叫 MARS，
05:39
Multi-Appendage Robotic System. So, it's a hexapod robot.
128
324000
3000
多附加機器人工程系統。是的，這是個六足機器人，
05:42
We developed our adaptive gait planner.
129
327000
2000
我們開發了自己的適應性步伐平台，
05:44
We actually have a very interesting payload on there.
130
329000
2000
實際上，我們放了一個很有趣的貨物在它上面。
05:46
The students like to have fun. And here you can see that it's
131
331000
2000
學生們喜歡有趣的東西。這裡你可以看到它
05:48
walking over unstructured terrain.
132
333000
3000
走過沒有結構可言的地形。
05:51
It's trying to walk on the coarse terrain,
133
336000
2000
它正嘗試穿過這粗糙的地形，
05:53
sandy area,
134
338000
2000
沙地，
05:55
but depending on the moisture content or the grain size of the sand
135
340000
5000
根據濕度大小或者沙子顆粒大小，
06:00
the foot's soil sinkage model changes.
136
345000
2000
它腳的泥土陷入模式會改變，
06:02
So, it tries to adapt its gait to successfully cross over these kind of things.
137
347000
4000
所以，它嘗試把它的步伐調整到能成功地穿過這些地方。
06:06
And also, it does some fun stuff, as can imagine.
138
351000
2000
因此，可以想像，它會做出一些有趣的動作，
06:08
We get so many visitors visiting our lab.
139
353000
3000
很多人會到我們的實驗室參觀，
06:11
So, when the visitors come, MARS walks up to the computer,
140
356000
2000
當有參觀者來的時候，MARS 會走到電腦前，
06:13
starts typing "Hello, my name is MARS."
141
358000
2000
開始打字：「嗨，我叫 MARS。」
06:15
Welcome to RoMeLa,
142
360000
2000
歡迎來到 RoMeLa，
06:17
the Robotics Mechanisms Laboratory at Virginia Tech.
143
362000
4000
在維吉尼亞理工的機器人工程實驗室。
06:21
This robot is an amoeba robot.
144
366000
2000
這是個阿米巴機器人，
06:23
Now, we don't have enough time to go into technical details,
145
368000
3000
們沒有時間詳細討論技術細節，
06:26
I'll just show you some of the experiments.
146
371000
2000
我將直接展示一些實驗給你們看。
06:28
So, this is some of the early feasibility experiments.
147
373000
2000
這是我們一些早期的可行性實驗。
06:30
We store potential energy to the elastic skin to make it move.
148
375000
4000
我們在它的彈性皮膚上儲存位能用以驅動它。
06:34
Or use an active tension cords to make it move
149
379000
2000
或者用一個彈簧索來讓它
06:36
forward and backward. It's called ChIMERA.
150
381000
3000
向前或者向後移動。它叫做 ChiMERA.
06:39
We also have been working with some scientists
151
384000
2000
我們也曾和一些來自賓大的
06:41
and engineers from UPenn
152
386000
2000
科學家和工程師合作
06:43
to come up with a chemically actuated version
153
388000
2000
開發了一個這阿米巴機器人的
06:45
of this amoeba robot.
154
390000
2000
化學能驅動版本。
06:47
We do something to something
155
392000
2000
我們這兒加點，那兒也加點。
06:49
And just like magic, it moves. The blob.
156
394000
6000
就像魔術一樣，結果，它動了，這小圓塊。
06:55
This robot is a very recent project. It's called RAPHaEL.
157
400000
2000
這機器人是一個非常新的計畫，叫 RAPHaEL，
06:57
Robotic Air Powered Hand with Elastic Ligaments.
158
402000
3000
有彈性韌帶的空氣驅動手機器人。
07:00
There are a lot of really neat, very good robotic hands out there in the market.
159
405000
4000
市面上有很多很好，很巧妙地機械手，
07:04
The problem is they're just too expensive, tens of thousands of dollars.
160
409000
4000
問題是，它們都過於昂貴，每個需要成千上萬的美金。
07:08
So, for prosthesis applications it's probably not too practical,
161
413000
2000
所以，應用在義肢上可能並不會太實用，
07:10
because it's not affordable.
162
415000
2000
因為大多數人負擔不起。
07:12
We wanted to go tackle this problem in a very different direction.
163
417000
4000
我們希望能從一個十分不同的角度來解決這個問題，
07:16
Instead of using electrical motors, electromechanical actuators,
164
421000
3000
我們不再用電動馬達和其它的機電驅動，
07:19
we're using compressed air.
165
424000
2000
我們使用壓縮空氣，
07:21
We developed these novel actuators for joints.
166
426000
2000
我們開發了這些新型的關節驅動器，
07:23
It is compliant. You can actually change the force,
167
428000
3000
它的兼容性很高。你可以簡單地
07:26
simply just changing the air pressure.
168
431000
2000
通過改變氣壓改變力度，
07:28
And it can actually crush an empty soda can.
169
433000
2000
它可以壓扁空的易開罐。
07:30
It can pick up very delicate objects like a raw egg,
170
435000
3000
也可以撿起像生雞蛋這樣易碎的物體。
07:33
or in this case, a lightbulb.
171
438000
3000
或像這邊，一個燈泡。
07:36
The best part, it took only $200 dollars to make the first prototype.
172
441000
4000
最妙的是，它只需要二百美金，就能做出第一台原型機。
07:40
This robot is actually a family of snake robots
173
445000
3000
這機器人其實屬於蛇形機器人系列，
07:43
that we call HyDRAS,
174
448000
2000
我們稱它為 HyDRAS，
07:45
Hyper Degrees-of-freedom Robotic Articulated Serpentine.
175
450000
2000
蛇形高自由度機器人。
07:47
This is a robot that can climb structures.
176
452000
3000
這機器人可以攀爬。
07:50
This is a HyDRAS's arm.
177
455000
2000
這是 HyDRAS 的手臂，
07:52
It's a 12 degrees of freedom robotic arm.
178
457000
2000
這是有 12 個自由度的機械手臂。
07:54
But the cool part is the user interface.
179
459000
2000
但最酷的還是使用者界面。
07:56
The cable over there, that's an optical fiber.
180
461000
3000
電纜在那兒，這是一條光纖，
07:59
And this student, probably the first time using it,
181
464000
2000
而這學生，大概是第一次用，
08:01
but she can articulate it many different ways.
182
466000
2000
但她可以通過不同方式清楚傳達她的意思。
08:03
So, for example in Iraq, you know, the war zone,
183
468000
3000
舉個例子，在伊拉克，你知道，是個戰場，
08:06
there is roadside bombs. Currently you send these
184
471000
2000
那裡路邊有炸彈。 目前你可以派這
08:08
remotely controlled vehicles that are armed.
185
473000
3000
武裝的遙控載具過去，
08:11
It takes really a lot of time and it's expensive
186
476000
2000
這真的會花費很多時間和金錢
08:13
to train the operator to operate this complex arm.
187
478000
4000
去訓練操縱員操作這複雜的機械手臂，
08:17
In this case it's very intuitive;
188
482000
2000
而這邊則是十分直覺的。
08:19
this student, probably his first time using it, doing very complex manipulation tasks,
189
484000
4000
這學生，大概是第一次用，正在進行十分複雜的操縱任務，
08:23
picking up objects and doing manipulation,
190
488000
2000
撿起物體，並進行操控，
08:25
just like that. Very intuitive.
191
490000
3000
像這樣，非常直觀。
08:30
Now, this robot is currently our star robot.
192
495000
2000
這個機器人是我們的明星級機器人。
08:32
We actually have a fan club for the robot, DARwIn:
193
497000
3000
我們的 DARwin 機器人有一個粉絲團，
08:35
Dynamic Anthropomorphic Robot with Intelligence.
194
500000
3000
人形動力智能機器人。
08:38
As you know, we are very interested in
195
503000
2000
如你所知，我們對人形機器人，
08:40
humanoid robot, human walking,
196
505000
2000
以及人類行走非常感興趣，
08:42
so we decided to build a small humanoid robot.
197
507000
2000
所以，我們打算做一個小型的人形機器人。
08:44
This was in 2004; at that time,
198
509000
2000
那是在 2004 年，
08:46
this was something really, really revolutionary.
199
511000
2000
在當時，這計畫非常、非常的有革命性。
08:48
This was more of a feasibility study:
200
513000
2000
這更像是一項可行性研究，
08:50
What kind of motors should we use?
201
515000
2000
我們應該用怎麼樣的馬達呢？
08:52
Is it even possible? What kinds of controls should we do?
202
517000
2000
這真的可行嗎？我們要如何控制呢？
08:54
So, this does not have any sensors.
203
519000
2000
沒有任何的感應器，
08:56
So, it's an open loop control.
204
521000
2000
開環控制，
08:58
For those who probably know, if you don't have any sensors
205
523000
2000
你們中一些人可能會知道，如果你沒有任何感應器，
09:00
and there are any disturbances, you know what happens.
206
525000
2000
而同時又有干擾的時候，會有什麽事情發生。
09:05
(Laughter)
207
530000
1000
（笑聲）
09:06
So, based on that success, the following year
208
531000
2000
所以，基於這些成就，後面一年
09:08
we did the proper mechanical design
209
533000
3000
我們做了適當的機械設計，
09:11
starting from kinematics.
210
536000
2000
從運動學開始。
09:13
And thus, DARwIn I was born in 2005.
211
538000
2000
最後，DARwin 一代在 2005 年誕生。
09:15
It stands up, it walks -- very impressive.
212
540000
2000
它站起來了，還能走了，十分讓人震驚。
09:17
However, still, as you can see,
213
542000
2000
儘管如此，你還是能看到，
09:19
it has a cord, umbilical cord. So, we're still using an external power source
214
544000
4000
它有條電線，臍帶，我們仍然使用外部電源
09:23
and external computation.
215
548000
2000
和外部的計算。
09:25
So, in 2006, now it's really time to have fun.
216
550000
4000
所以，到了 2006 年，是時候來點有趣的了。
09:29
Let's give it intelligence. We give it all the computing power it needs:
217
554000
3000
我們給了它智能，我們給了它所有它需要的計算能力，
09:32
a 1.5 gigahertz Pentium M chip,
218
557000
2000
1.5千兆赫的奔騰M晶片，
09:34
two FireWire cameras, rate gyros, accelerometers,
219
559000
2000
兩個火線攝影機，八個陀螺儀，加速度計，
09:36
four force sensors on the foot, lithium polymer batteries.
220
561000
3000
四個腳上的扭矩感應器，鋰電池。
09:39
And now DARwIn II is completely autonomous.
221
564000
4000
現在的 DARwin 二代完全自動化了，
09:43
It is not remote controlled.
222
568000
2000
它不再是遙控的了。
09:45
There are no tethers. It looks around, searches for the ball,
223
570000
3000
沒有外部繫繩。看，它朝四處看，尋找那個球，
09:48
looks around, searches for the ball, and it tries to play a game of soccer,
224
573000
3000
不停地看，搜索那個球，而且它嘗試著去踢足球，
09:51
autonomously: artificial intelligence.
225
576000
3000
這就是自動化的，人工智能。
09:54
Let's see how it does. This was our very first trial,
226
579000
3000
讓我們來看它是怎樣做到的。這是我們第一次的嘗試，
09:57
and... Spectators (Video): Goal!
227
582000
5000
影片：進了！
10:03
Dennis Hong: So, there is actually a competition called RoboCup.
228
588000
3000
這其實是一個叫作 RoboCup 的比賽。
10:06
I don't know how many of you have heard about RoboCup.
229
591000
2000
我不知道你們有多少人聽過 RoboCup，
10:08
It's an international autonomous robot soccer competition.
230
593000
5000
這是一個國際性的自動化機器人足球比賽。
10:13
And the goal of RoboCup, the actual goal is,
231
598000
3000
RoboCup 真正的目標是，
10:16
by the year 2050
232
601000
2000
到 2050 年，
10:18
we want to have full size, autonomous humanoid robots
233
603000
3000
我們會有全尺寸、自動化的人形機器人
10:21
play soccer against the human World Cup champions
234
606000
4000
足球隊與人類世界杯冠軍對壘，
10:25
and win.
235
610000
2000
而且，還要打贏人類。
10:27
It's a true actual goal. It's a very ambitious goal,
236
612000
2000
這是個千真萬確的目標，也是個十分有野心的目標，
10:29
but we truly believe that we can do it.
237
614000
2000
但我們確信，我們可以做到。
10:31
So, this is last year in China.
238
616000
3000
這是去年在中國，
10:34
We were the very first team in the United States that qualified
239
619000
2000
我們是美國第一個獲得
10:36
in the humanoid RoboCup competition.
240
621000
2000
人形機器人競賽參賽資格的隊伍。
10:38
This is this year in Austria.
241
623000
3000
而今年，在奧地利。
10:41
You're going to see the action, three against three,
242
626000
2000
你們即將看到這比賽，是三對三，
10:43
completely autonomous.
243
628000
2000
完全地自動化。
10:45
There you go. Yes!
244
630000
2000
進了！耶！
10:48
The robots track and they
245
633000
2000
機器人跟著球跑，而且它們能有組織地踢，
10:50
team play amongst themselves.
246
635000
3000
完全由它們自己組織。
10:53
It's very impressive. It's really a research event
247
638000
2000
這讓人十分驚訝。這其實是一個包裝在
10:55
packaged in a more exciting competition event.
248
640000
4000
讓人興奮地比賽活動裡的研究項目。
10:59
What you see over here, this is the beautiful
249
644000
2000
你在這裡看到的，真是漂亮，
11:01
Louis Vuitton Cup trophy.
250
646000
2000
Louis Vuitton Cup 獎盃。
11:03
So, this is for the best humanoid,
251
648000
2000
這是給最佳人形機器人的，
11:05
and we would like to bring this for the very first time, to the United States
252
650000
2000
我們很高興能將這個獎盃首次捧回美國，
11:07
next year, so wish us luck.
253
652000
2000
下一年，希望我們也能夠如此幸運。
11:09
(Applause)
254
654000
2000
謝謝。
11:11
Thank you.
255
656000
3000
（掌聲）
11:14
DARwIn also has a lot of other talents.
256
659000
2000
DARwin 還有很多其它技能。
11:16
Last year it actually conducted the Roanoke Symphony Orchestra
257
661000
3000
去年，在假日音樂會上，
11:19
for the holiday concert.
258
664000
3000
它指揮了 Roanoke 交響樂團。
11:22
This is the next generation robot, DARwIn IV,
259
667000
3000
這是下一代機器人，DARwin 四代，
11:25
but smarter, faster, stronger.
260
670000
3000
更聰明，更快，更強。
11:28
And it's trying to show off its ability:
261
673000
2000
它正炫耀它的本領。
11:30
"I'm macho, I'm strong.
262
675000
3000
「看我強壯又威武。」
11:33
I can also do some Jackie Chan-motion,
263
678000
3000
「我也可以做一些成龍
11:36
martial art movements."
264
681000
3000
電影裡的武術動作啊」
11:39
(Laughter)
265
684000
2000
（笑聲）
11:41
And it walks away. So, this is DARwIn IV.
266
686000
2000
然後，它走掉了，這，就是 DARwin 四代,
11:43
And again, you'll be able to see it in the lobby.
267
688000
2000
你們會有機會在門廳再一次看到它的。
11:45
We truly believe this is going to be the very first running
268
690000
2000
我們真的相信這會是美國第一個
11:47
humanoid robot in the United States. So, stay tuned.
269
692000
3000
能奔跑的人型機器人，所以拭目以待吧。
11:50
All right. So I showed you some of our exciting robots at work.
270
695000
3000
我已經展示一些我們工作上所製作的激動人心的機器人。
11:53
So, what is the secret of our success?
271
698000
3000
那麼我們成功的秘密是什麽呢?
11:56
Where do we come up with these ideas?
272
701000
2000
我們這些創意從哪兒來的呢?
11:58
How do we develop these kinds of ideas?
273
703000
2000
我們是如何開發這些創意的呢？
12:00
We have a fully autonomous vehicle
274
705000
2000
我們有一輛全自動的車，
12:02
that can drive into urban environments. We won a half a million dollars
275
707000
2000
這車可以開到城市里。結果我們在 DARPA
12:04
in the DARPA Urban Challenge.
276
709000
2000
城市挑戰賽中贏了 50 萬美金。
12:06
We also have the world's very first
277
711000
2000
我們也研發了世界上首台
12:08
vehicle that can be driven by the blind.
278
713000
2000
能被盲人駕駛的汽車。
12:10
We call it the Blind Driver Challenge, very exciting.
279
715000
2000
我們叫這做盲人司機挑戰賽，非常好玩，
12:12
And many, many other robotics projects I want to talk about.
280
717000
4000
還有很多很多其它我想介紹的機器人計畫。
12:16
These are just the awards that we won in 2007 fall
281
721000
2000
這些只是我們在 2007 年秋天贏得的獎項，
12:18
from robotics competitions and those kinds of things.
282
723000
3000
是機器人比賽之類的。
12:21
So, really, we have five secrets.
283
726000
2000
我們有五個秘訣。
12:23
First is: Where do we get inspiration?
284
728000
2000
首先是我們的啓發從哪兒來？
12:25
Where do we get this spark of imagination?
285
730000
2000
我們從哪兒得到靈機一動的瞬間？
12:27
This is a true story, my personal story.
286
732000
3000
這是一個真實的故事，我自己的故事。
12:30
At night when I go to bed, 3 - 4 a.m. in the morning,
287
735000
2000
當我晚上上床時，凌晨三、四點吧，
12:32
I lie down, close my eyes, and I see these lines and circles
288
737000
3000
我躺在床上，閤上眼睛，我看到這些線條和圓圈，
12:35
and different shapes floating around.
289
740000
2000
還有其它的形狀一直在周圍浮動，
12:37
And they assemble, and they form these kinds of mechanisms.
290
742000
3000
然後，它們組合起來，形成這樣的機械組件。
12:40
And then I think, "Ah this is cool."
291
745000
2000
我覺得，「啊，這超酷的。」
12:42
So, right next to my bed I keep a notebook,
292
747000
2000
我一般都在我床頭準備一本筆記本，
12:44
a journal, with a special pen that has a light on it, LED light,
293
749000
3000
一個有著一支帶 LED 燈的筆在上面的筆記本，
12:47
because I don't want to turn on the light and wake up my wife.
294
752000
2000
因為我不想打開燈吵醒我妻子。
12:49
So, I see this, scribble everything down, draw things,
295
754000
2000
就這樣，我記下所有東西，畫好它，
12:51
and I go to bed.
296
756000
2000
再上床睡覺。
12:53
Every day in the morning,
297
758000
2000
每天早晨
12:55
the first thing I do before my first cup of coffee,
298
760000
2000
我做的第一件事，不是給自己倒杯咖啡，
12:57
before I brush my teeth, I open my notebook.
299
762000
2000
不是刷牙，而是打開我的筆記本。
12:59
Many times it's empty,
300
764000
2000
很多時候，它是空的，
13:01
sometimes I have something there -- if something's there, sometimes it's junk --
301
766000
2000
有時候，有東西了，卻是垃圾，
13:03
but most of the time I can't even read my handwriting.
302
768000
3000
更多時候，我不能看懂我自己寫的東西。
13:06
And so, 4 am in the morning, what do you expect, right?
303
771000
3000
而且，凌晨四點，你還能要求什麽呢？對吧？
13:09
So, I need to decipher what I wrote.
304
774000
2000
所以，我需要幫我所寫的東西解碼。
13:11
But sometimes I see this ingenious idea in there,
305
776000
3000
但是，有時候，我能看到一些絕妙的主意在這裡面，
13:14
and I have this eureka moment.
306
779000
2000
於是，靈光一閃，
13:16
I directly run to my home office, sit at my computer,
307
781000
2000
我直接跑到我家裡的辦公室，坐在我電腦旁，
13:18
I type in the ideas, I sketch things out
308
783000
2000
把這些想法打進電腦，把它們理清楚，
13:20
and I keep a database of ideas.
309
785000
3000
我有個點子資料庫。
13:23
So, when we have these calls for proposals,
310
788000
2000
所以，當我們有提案的需求時，
13:25
I try to find a match between my
311
790000
2000
我會嘗試在我的資料庫裡
13:27
potential ideas
312
792000
2000
找到匹配的可能想法
13:29
and the problem. If there is a match we write a research proposal,
313
794000
2000
和問題，如果我們找到匹配的，我們會寫個研究提案，
13:31
get the research funding in, and that's how we start our research programs.
314
796000
4000
找到研究資金，然後開始我們的研究項目。
13:35
But just a spark of imagination is not good enough.
315
800000
3000
但只有靈機一動時候的創意是不足夠的。
13:38
How do we develop these kinds of ideas?
316
803000
2000
我們怎樣改進這些創意呢？
13:40
At our lab RoMeLa, the Robotics and Mechanisms Laboratory,
317
805000
3000
在我們的 RoMeLa，機器人機械實驗室，
13:43
we have these fantastic brainstorming sessions.
318
808000
3000
我們有超棒的腦力激盪時間，
13:46
So, we gather around, we discuss about problems
319
811000
2000
我們會聚在一起，討論有關問題
13:48
and social problems and talk about it.
320
813000
2000
和社會問題，並分析它們。
13:50
But before we start we set this golden rule.
321
815000
3000
在我們開始之前，我們定下了鐵則，
13:53
The rule is:
322
818000
2000
規則是：
13:55
Nobody criticizes anybody's ideas.
323
820000
3000
不可以批評別人的想法、
13:58
Nobody criticizes any opinion.
324
823000
2000
不可以批評任何觀點、
14:00
This is important, because many times students, they fear
325
825000
2000
這很重要，因為很多時候，學生們會因為在乎
14:02
or they feel uncomfortable how others might think
326
827000
3000
別人對自己的觀點和思想的想法
14:05
about their opinions and thoughts.
327
830000
2000
而害怕或者感到不舒服
14:07
So, once you do this, it is amazing
328
832000
2000
當我們實施了這措施以後，我們很驚訝地發現
14:09
how the students open up.
329
834000
2000
學生們開始暢所欲言了。
14:11
They have these wacky, cool, crazy, brilliant ideas, and
330
836000
3000
他們有很多怪誕的，但是很酷很瘋狂的妙點子，
14:14
the whole room is just electrified with creative energy.
331
839000
3000
整個房間都被創意能量激發了，
14:17
And this is how we develop our ideas.
332
842000
3000
這就是我們改進想法的方法。
14:20
Well, we're running out of time. One more thing I want to talk about is,
333
845000
3000
我們的時間不多了，我還想說的一件事就是
14:23
you know, just a spark of idea and development is not good enough.
334
848000
4000
你知道，只有瞬間的靈光一閃和改進是不夠的。
14:27
There was a great TED moment,
335
852000
2000
曾有一個很好的 TED 演講，
14:29
I think it was Sir Ken Robinson, was it?
336
854000
3000
是 Ken Robinson 爵士，對吧？
14:32
He gave a talk about how education
337
857000
2000
他的演講是有關教育和學校
14:34
and school kills creativity.
338
859000
2000
如何扼殺創意。
14:36
Well, actually, there are two sides to the story.
339
861000
3000
其實這個故事有兩面，
14:39
So, there is only so much one can do
340
864000
3000
對於直率的想法、
14:42
with just ingenious ideas
341
867000
2000
創造力和好的工程直覺
14:44
and creativity and good engineering intuition.
342
869000
3000
一個人能做的就這麼點兒。
14:47
If you want to go beyond a tinkering,
343
872000
2000
如果你不只想做個貪玩的孩子，
14:49
if you want to go beyond a hobby of robotics
344
874000
2000
如果你不只想讓機器人成為你的興趣，
14:51
and really tackle the grand challenges of robotics
345
876000
3000
而是真的想在這方面通過進行精確的
14:54
through rigorous research
346
879000
2000
研究來挑戰自我
14:56
we need more than that. This is where school comes in.
347
881000
3000
我們需要更多，這才是學校的目的。
14:59
Batman, fighting against bad guys,
348
884000
3000
蝙蝠俠，和壞人鬥爭，
15:02
he has his utility belt, he has his grappling hook,
349
887000
2000
他有實用的皮帶，有強大抓力的攀牆鉤，
15:04
he has all different kinds of gadgets.
350
889000
2000
有各種不同的小配件。
15:06
For us roboticists, engineers and scientists,
351
891000
2000
對於我們機器人學家，工程師和科學家，
15:08
these tools, these are the courses and classes you take in class.
352
893000
5000
這些工具就是你在課堂上學的課程，
15:13
Math, differential equations.
353
898000
2000
數學、微分方程、
15:15
I have linear algebra, science, physics,
354
900000
2000
線性代數、科學、物理，
15:17
even nowadays, chemistry and biology, as you've seen.
355
902000
3000
今天還有，化學和生物。
15:20
These are all the tools that we need.
356
905000
2000
這些就是我們需要的工具。
15:22
So, the more tools you have, for Batman,
357
907000
2000
對於蝙蝠俠來說，有更多的工具
15:24
more effective at fighting the bad guys,
358
909000
2000
對付壞人的時候會更有效，
15:26
for us, more tools to attack these kinds of big problems.
359
911000
4000
對於我們，就是用更多的工具對付這些大難題。
15:30
So, education is very important.
360
915000
3000
所以教育十分重要。
15:33
Also, it's not about that,
361
918000
2000
當然也不止這麼簡單，
15:35
only about that. You also have to work really, really hard.
362
920000
2000
最重要的，還是要努力學習。
15:37
So, I always tell my students,
363
922000
2000
我經常告訴我的學生
15:39
"Work smart, then work hard."
364
924000
2000
先求明智地工作，再認真地工作，
15:41
This picture in the back this is 3 a.m. in the morning.
365
926000
3000
這後面的照片是在凌晨三點拍的。
15:44
I guarantee if you come to your lab at 3 - 4 am
366
929000
2000
我能保證如果你凌晨三、四點來我們實驗室，
15:46
we have students working there,
367
931000
2000
還能看到有學生在那工作，
15:48
not because I tell them to, but because we are having too much fun.
368
933000
3000
不是因為我叫他們這樣，而是他們覺得這樣很好玩。
15:51
Which leads to the last topic:
369
936000
2000
這就把我們帶到最後一個話題，
15:53
Do not forget to have fun.
370
938000
2000
不要忘記給自己點樂趣，
15:55
That's really the secret of our success, we're having too much fun.
371
940000
3000
這是我們成功最重要的秘密，我們真的覺得很好玩，
15:58
I truly believe that highest productivity comes when you're having fun,
372
943000
3000
我真的相信最高的效率來自於當你找到樂趣的時候。
16:01
and that's what we're doing.
373
946000
2000
這就是我們所做的成果。
16:03
There you go. Thank you so much.
374
948000
2000
謝謝。
16:05
(Applause)
375
950000
5000
（掌聲）
ABOUT THE SPEAKER
Dennis Hong - RoboticistDennis Hong is the founder and director of RoMeLa -- a Virginia Tech robotics lab that has pioneered several breakthroughs in robot design and engineering.
Why you should listen
As director of a groundbreaking robotics lab, Dennis Hong guides his team of students through projects on robot locomotion and mechanism design, creating award-winning humanoid robots like DARwIn (Dynamic Anthropomorphic Robot with Intelligence). His team is known as RoMeLa (Robotics & Mechanisms Laboratory) and operates at Virginia Tech.
Hong has also pioneered various innovations in soft-body robots, using a “whole-skin locomotion” as inspired by amoebae. Marrying robotics with biochemistry, he has been able to generate new types of motion with these ingenious forms. For his contributions to the field, Hong was selected as a NASA Summer Faculty Fellow in 2005, given the CAREER award by the National Science Foundation in 2007 and in 2009, named as one of Popular Science's Brilliant 10. He is also a gourmet chef and a magician, performing shows for charity and lecturing on the science of magic.
Dennis Hong | Speaker | TED.com