ABOUT THE SPEAKER
Chris Gerdes - Mechanical engineer
An autonomous car may seem like a thing of the distant future, but mechanical engineer Chris Gerdes is racing to make it a reality today.

Why you should listen

Imagine a car that can drive itself -- that with the push of a button can get you home safely when you’re too tired to drive or have had a night of one too many drinks. Dr. Chris Gerdes , the Director of the Center for Automotive Research at Stanford (conveniently acronymed CARS), and his team are developing a robotic race car, capable of driving at outrageous speeds while avoiding every possible accident. Gerdes’ research focuses on the development of driver assistance systems for collision avoidance, as well as on new combustion processes for engines.

Prior to teaching at Stanford, Gerdes was the project leader for vehicle dynamics at the Vehicle Systems Technology Center of Daimler-Benz Research and Technology North America. His work at Daimler focused on safety analysis.

More profile about the speaker
Chris Gerdes | Speaker | TED.com
TEDxStanford

Chris Gerdes: The future race car -- 150mph, and no driver

基斯•吉迪斯: 未來的賽車 -- 時速150英里,沒有駕駛員

Filmed:
806,444 views

自動汽車快要到來了 -- 它們的操控會比你優勝。基斯•吉迪斯講述他和他的團隊怎樣研發可以走時速150英里而避免任何意外的機動賽車。同時,在研究專業賽車手的大腦電波的過程中,吉迪斯說他領略到一種對專業賽車手的本能反應的欣賞。(攝於 TEDx史丹福大學)
- Mechanical engineer
An autonomous car may seem like a thing of the distant future, but mechanical engineer Chris Gerdes is racing to make it a reality today. Full bio

Double-click the English transcript below to play the video.

00:16
So, how many許多 of you have ever
0
506
1407
在你們當中有多少人曾經
00:17
gotten得到 behind背後 the wheel of a car汽車
1
1913
1655
在你根本不應該駕駛的狀況下
00:19
when you really shouldn't不能 have been driving主動?
2
3568
5687
坐在駕駛者的位上呢?
00:25
Maybe you're out on the road for a long day,
3
9255
1905
也許你曾經一天開車很久了
00:27
and you just wanted to get home.
4
11160
1490
當下你只想回家
00:28
You were tired, but you felt you could drive駕駛 a few少數 more miles英里.
5
12650
2647
你非常累,但是你覺得你還可以再開幾哩路
00:31
Maybe you thought,
6
15297
1199
當下你也許在想,
00:32
I've had less to drink than everybody每個人 else其他,
7
16496
2017
我喝的酒比其他人都還少
00:34
I should be the one to go home.
8
18513
1736
應是我開車回家吧
00:36
Or maybe your mind心神 was just entirely完全 elsewhere別處.
9
20249
4591
又或者你的注意力完全不在開車這件事上了
00:40
Does this sound聲音 familiar to you?
10
24840
1454
這些想法聽起來很熟悉吧?
00:42
Now, in those situations情況, wouldn't不會 it be great
11
26294
2898
現在,在剛剛所說的那些狀況下,
你應該會感到以下的新發明很吸引吧...
00:45
if there was a button按鍵 on your dashboard儀表板
12
29192
1593
如果有一個按鈕在你的儀表板上,按一按
00:46
that you could push, and the car汽車 would get you home safely安然?
13
30785
6343
車子會自動地把你安全地送回家
00:53
Now, that's been the promise諾言 of the self-driving自駕車 car汽車,
14
37128
2293
這是自動導航汽車的承諾
00:55
the autonomous自主性 vehicle車輛, and it's been the dream夢想
15
39421
2627
一輛自主的汽車,這是從一九三九年
00:57
since以來 at least最小 1939, when General一般 Motors汽車 showcased展出
16
42048
3249
由通用汽車公司
01:01
this idea理念 at their Futurama飛出個未來 booth at the World's世界 Fair公平.
17
45297
3302
在世界博覽會“飛出個未來"攤位中展示他們的想法開始後,很多人對汽車的寄望
01:04
Now, it's been one of those dreams
18
48599
1943
現在, 這些夢想
01:06
that's always seemed似乎 about 20 years年份 in the future未來.
19
50542
4214
好像永遠要在二十年後才有希望實現
01:10
Now, two weeks ago, that dream夢想 took a step forward前鋒,
20
54756
2683
現在, 就在兩星期前, 這個夢想向前邁進了一步
01:13
when the state of Nevada內華達 granted理所當然 Google's谷歌的 self-driving自駕車 car汽車
21
57439
3265
當美國內華達州對 Google (谷歌) 的自動導航汽車發出
01:16
the very first license執照 for an autonomous自主性 vehicle車輛,
22
60704
3600
首張針對自動汽車的行車證
01:20
clearly明確地 establishing建立 that it's legal法律 for them
23
64304
2245
明確確立法洌上容許這輛車
01:22
to test測試 it on the roads道路 in Nevada內華達.
24
66549
1810
可以在內華達州的道路上進行測試
01:24
Now, California's加州 considering考慮 similar類似 legislation立法,
25
68359
3727
最近, 加州正考慮仿效類似的立法工作
01:27
and this would make sure that the autonomous自主性 car汽車
26
72086
2408
這樣一來, 他們就可以確保那輛自動汽車
01:30
is not one of those things that has to stay in Vegas拉斯維加斯.
27
74494
2977
不會只在內華達州出現
01:33
(Laughter笑聲)
28
77471
2096
(笑聲)
01:35
Now, in my lab實驗室 at Stanford斯坦福, we've我們已經 been working加工 on
29
79567
3784
在我的史丹福大學實驗室裏, 我們也在著手研究
01:39
autonomous自主性 cars汽車 too, but with a slightly different不同 spin
30
83351
3487
自動汽車, 但是從一個稍為不同的角度
01:42
on things. You see, we've我們已經 been developing發展 robotic機器人 race種族 cars汽車,
31
86838
4248
去看這件事。你看, 我們正在研發機動賽車
01:46
cars汽車 that can actually其實 push themselves他們自己 to the very limits範圍
32
91086
4120
它們可以把自己的物理表現
01:51
of physical物理 performance性能.
33
95206
2240
推至極限
01:53
Now, why would we want to do such這樣 a thing?
34
97446
2613
為甚麼我們會想做這件事呢?
01:55
Well, there's two really good reasons原因 for this.
35
100059
2100
嗯, 這裏有兩個非常好的原因
01:58
First, we believe that before people turn over control控制
36
102159
3959
第一, 我們相信, 在人類把控制權
02:02
to an autonomous自主性 car汽車, that autonomous自主性 car汽車 should be
37
106118
2834
交給自動汽車之前, 那自動汽車的表現至少要
02:04
at least最小 as good as the very best最好 human人的 drivers司機.
38
108952
3254
能夠與最好的人類駕駛者匹敵
02:08
Now, if you're like me, and the other 70 percent百分 of the population人口
39
112206
3305
如果你跟我一樣, 跟百分之七十的人口一樣
02:11
who know that we are above-average高於平均水平 drivers司機,
40
115511
2193
都知道我們比一般駕駛者優勝
02:13
you understand理解 that's a very high bar酒吧.
41
117704
3175
你會明白這對於自動汽車來說是很高的門檻
02:16
There's another另一個 reason原因 as well.
42
120879
2392
還有另一個原因
02:19
Just like race種族 car汽車 drivers司機 can use all of the friction摩擦
43
123271
3576
就像賽車手能夠充份運用
02:22
between之間 the tire and the road,
44
126847
1280
輪胎與路面之間的摩擦力
02:24
all of the car's汽車 capabilities功能 to go as fast快速 as possible可能,
45
128127
3177
加上充份運用車子的性能去達致最高速
02:27
we want to use all of those capabilities功能 to avoid避免
46
131304
3345
我們想運用那些性質
02:30
any accident事故 we can.
47
134649
1588
去減少任何意外的發生
02:32
Now, you may可能 push the car汽車 to the limits範圍
48
136237
2050
有時你使車子超過所能承受的極限
02:34
not because you're driving主動 too fast快速,
49
138287
1967
未必因為你把車子開得太快
02:36
but because you've hit擊中 an icy冷冰冰 patch補丁 of road,
50
140254
2160
而是因為你碰到地上的冰塊
02:38
conditions條件 have changed.
51
142414
1704
以致於狀況有所改變。
02:40
In those situations情況, we want a car汽車
52
144118
2761
在那些情況, 我們想要有
02:42
that is capable enough足夠 to avoid避免 any accident事故
53
146879
3720
只要是物理上能被避免的
02:46
that can physically物理 be avoided避免.
54
150599
2678
就能購避免任何意外發生的一輛車。
02:49
I must必須 confess承認, there's kind of a third第三 motivation動機 as well.
55
153277
4267
我必須承認, 我們有第三個動機
02:53
You see, I have a passion for racing賽跑.
56
157544
2256
你看, 我有一股對賽車的熱誠
02:55
In the past過去, I've been a race種族 car汽車 owner所有者,
57
159800
2764
以前, 我曾經是一輛賽車車主
02:58
a crew船員 chief首席 and a driving主動 coach教練,
58
162564
2555
一個車隊的領隊及駕駛教練
03:01
although雖然 maybe not at the level水平 that you're currently目前 expecting期待.
59
165119
3855
雖然不是你們想像中的那個層次
03:04
One of the things that we've我們已經 developed發達 in the lab實驗室 --
60
168974
2704
我們在實驗室裏研發的其中一個項目--
03:07
we've我們已經 developed發達 several一些 vehicles汽車 --
61
171678
1704
我們研發過幾款汽車--
03:09
is what we believe is the world's世界 first
62
173382
2235
是我們認為是世界上第一輛
03:11
autonomously自主 drifting漂流 car汽車.
63
175617
2365
會飄移的自動汽車
03:13
It's another另一個 one of those categories類別
64
177982
2513
這是一個比較少
03:16
where maybe there's not a lot of competition競爭.
65
180495
2423
競爭對手的研究範疇
03:18
(Laughter笑聲)
66
182918
1408
(笑聲)
03:20
But this is P1. It's an entirely完全 student-built學生建 electric電動 vehicle車輛,
67
184326
3822
這是 P1。一個完全由學生建立的電動汽車
03:24
which哪一個 through通過 using運用 its rear-wheel後輪 drive駕駛
68
188148
2078
它用它的尾輪
03:26
and front-wheel前輪 steer-by-wire轉向的線控
69
190226
1565
以及電子操控的前輪
03:27
can drift漂移 around corners角落.
70
191791
2067
可以在車道的彎位飄移
03:29
It can get sideways側身 like a rally團結 car汽車 driver司機,
71
193858
2200
它可以像拉力賽中的賽車手高速轉向
03:31
always able能夠 to take the tightest嚴格的 curve曲線,
72
196058
1715
永遠能夠在最窄的角度行走
03:33
even on slippery, changing改變 surfaces,
73
197773
3304
即使賽道濕滑、路面情況轉變
03:36
never spinning紡織 out.
74
201077
1616
它都不會滑出賽道
03:38
We've我們已經 also worked工作 with Volkswagen大眾汽車 Oracle神諭,
75
202693
2368
我們也跟福特汽車合作
03:40
on Shelley雪萊, an autonomous自主性 race種族 car汽車 that has raced賽跑
76
205061
3424
用它的Shelley 型號汽車, 一款自動賽車
03:44
at 150 miles英里 an hour小時 through通過 the Bonneville博納維爾 Salt Flats公寓,
77
208485
3070
以時速150 英里穿越美國猶他州邦納維爾大鹽湖附近的平原
03:47
gone走了 around ThunderhillThunderhill Raceway賽道 Park公園 in the sun太陽,
78
211555
4471
在太陽底下橫過 Thunderhill 賽車場
03:51
the wind and the rain,
79
216026
2639
也在大風和雨中行走過該處
03:54
and navigated導航 the 153 turns and 12.4 miles英里
80
218665
5018
以導航系統引領汽車在科羅拉多州的 Pikes Peak Hill Climb 賽道上
03:59
of the Pikes派克 Peak Hill爬坡道 Climb route路線
81
223683
1562
轉153 個圈及走12.4 英里
04:01
in Colorado科羅拉多州 with nobody沒有人 at the wheel.
82
225245
3473
汽車是無人駕駛的
04:04
(Laughter笑聲)
83
228718
1448
(笑聲)
04:06
(Applause掌聲)
84
230166
5566
(掌聲)
04:11
I guess猜測 it goes without saying that we've我們已經 had a lot of fun開玩笑
85
235732
3279
我毋須多說,這些經歷為我們
04:14
doing this.
86
239011
1304
帶來無窮的樂趣
04:16
But in fact事實, there's something else其他 that we've我們已經 developed發達
87
240315
3657
但實際上,我們在研發了
04:19
in the process處理 of developing發展 these autonomous自主性 cars汽車.
88
243972
3055
這些自動汽車的過程中,體驗到一些事情
04:22
We have developed發達 a tremendous巨大 appreciation升值
89
247027
3871
我們培養出一種對人類賽車手
04:26
for the capabilities功能 of human人的 race種族 car汽車 drivers司機.
90
250898
3817
的能力的高度欣賞
04:30
As we've我們已經 looked看著 at the question of how well do these cars汽車 perform演出,
91
254715
4345
當我們看到這些自動汽車的表現時
04:34
we wanted to compare比較 them to our human人的 counterparts同行.
92
259060
3279
我們想將它們與人類賽車手作比較
04:38
And we discovered發現 their human人的 counterparts同行 are amazing驚人.
93
262339
5680
我們發現人類賽車手的表現是多麼令人讚歎的
04:43
Now, we can take a map地圖 of a race種族 track跟踪,
94
268019
4023
如果我們拿到賽道的地圖
04:47
we can take a mathematical數學的 model模型 of a car汽車,
95
272042
2370
我們可以為自動汽車製作一個數學模型
04:50
and with some iteration迭代, we can actually其實 find
96
274412
2903
經過幾個回合的修定,我們的確可以找到
04:53
the fastest最快的 way around that track跟踪.
97
277315
1625
穿越賽道的最快路徑
04:54
We line that up with data數據 that we record記錄
98
278940
2533
我們將這些記錄得來的數據
04:57
from a professional專業的 driver司機,
99
281473
1433
與專業的賽車手比較
04:58
and the resemblance相似 is absolutely絕對 remarkable卓越.
100
282906
4107
結果兩者的吻合實在值得靈人注意
05:02
Yes, there are subtle微妙 differences分歧 here,
101
287013
3212
是的,兩者有少許的差異
05:06
but the human人的 race種族 car汽車 driver司機 is able能夠 to go out
102
290225
3127
但人類賽車手能夠出賽
05:09
and drive駕駛 an amazingly令人驚訝 fast快速 line,
103
293352
2335
然後奇妙的高速行駛
05:11
without the benefit效益 of an algorithm算法 that compares比較
104
295687
2330
卻毋須任何演算法去計算
05:13
the trade-off交易 between之間 going as fast快速 as possible可能
105
298017
2608
車子以最高速轉向
05:16
in this corner, and shaving a little bit of time
106
300625
2037
以及不在直路上行走所損失的時間
05:18
off of the straight直行 over here.
107
302662
1902
兩者之間的取捨
05:20
Not only that, they're able能夠 to do it lap膝部
108
304564
3457
不單如此,他們能夠在一在一圈
05:23
after lap膝部 after lap膝部.
109
308021
2375
又一圈... 又一圈後做出同樣的決定
05:26
They're able能夠 to go out and consistently始終如一 do this,
110
310396
2912
他們在賽道上的表現非常穩定
05:29
pushing推動 the car汽車 to the limits範圍 every一切 single time.
111
313308
4128
每一次都把車子的性能發揮到極限
05:33
It's extraordinary非凡 to watch.
112
317436
3169
這的確令人歎為觀止
05:36
You put them in a new car汽車,
113
320605
2066
你把他們放到一輛新車裏
05:38
and after a few少數 laps, they've他們已經 found發現 the fastest最快的 line in that car汽車,
114
322671
3902
跑幾個圈以後,他們就能夠為那新車找出最快的線道
05:42
and they're off to the races比賽.
115
326573
3877
然後他們就準備好去比賽
05:46
It really makes品牌 you think,
116
330450
1146
你真的會覺得
05:47
we'd星期三 love to know what's going on inside their brain.
117
331596
4871
我們是多麼渴望知道他們的腦袋是如何運作的
05:52
So as researchers研究人員, that's what we decided決定 to find out.
118
336467
4541
作為研究人員,這是我們決定追尋的問題
05:56
We decided決定 to instrument儀器 not only the car汽車,
119
341008
1812
我們不但決定裝備那輛車子
05:58
but also the race種族 car汽車 driver司機,
120
342820
2495
還要裝備那個駕駛者
06:01
to try to get a glimpse一瞥 into what was going on
121
345315
2769
去嘗試了解,在他們做出這些行為的時候
06:03
in their head as they were doing this.
122
348084
2186
他們的腦袋會產生怎麼的變化
06:06
Now, this is Dr博士. Lene琳恩 HarbottHarbott applying應用 electrodes電極
123
350270
3950
蘭尼•哈博 (Dr. Lene Harbott) 博士把電極
06:10
to the head of John約翰 Morton莫頓.
124
354220
1232
附在約翰•摩頓 (John Morton) 的頭上
06:11
John約翰 Morton莫頓 is a former前任的 Can-Am能是 and IMSAIMSA driver司機,
125
355452
2989
約翰•摩頓是加拿大-美國挑戰盃及國際賽車運動協會賽事的車手
06:14
who's誰是 also a class champion冠軍 at Le Mans.
126
358441
1800
亦是利曼24小时耐力赛的分組冠軍
06:16
Fantastic奇妙 driver司機, and very willing願意 to put up with graduate畢業 students學生們
127
360241
3496
這個優異的車手,而且非常願意和研究生合作
06:19
and this sort分類 of research研究.
128
363737
1855
及參與這類研究
06:21
She's putting electrodes電極 on his head
129
365592
2672
她正在把電極放到他的頭上
06:24
so that we can monitor監控 the electrical電動 activity活動
130
368264
2112
使我們能在約翰在賽道上飛馳的時候
06:26
in John's約翰 brain as he races比賽 around the track跟踪.
131
370376
2832
監察他的大腦活動
06:29
Now, clearly明確地 we're not going to put a couple一對 of electrodes電極 on his head
132
373208
3195
明顯地,我們不能以這幾支附在他頭上的電極
06:32
and understand理解 exactly究竟 what all of his thoughts思念 are on the track跟踪.
133
376403
3270
完全了解他在賽道的時候的思想
06:35
However然而, neuroscientists神經學家 have identified確定 certain某些 patterns模式
134
379673
3407
但是,神經系統科學家已經認定了幾個不同的規律
06:38
that let us tease out some very important重要 aspects方面 of this.
135
383080
3761
為我們提供了幾個非常重要的線索
06:42
For instance, the resting休息 brain
136
386841
1847
例如,大腦靜止的時候
06:44
tends趨向 to generate生成 a lot of alphaα waves波浪.
137
388688
2155
傾向產生很多 alpha 波段
06:46
In contrast對比, thetaTHETA waves波浪 are associated相關 with
138
390843
3752
相對地,theta 波段與
06:50
a lot of cognitive認知 activity活動, like visual視覺 processing處理,
139
394595
3184
很多認知活動有關,例如視覺訊息處理
06:53
things where the driver司機 is thinking思維 quite相當 a bit.
140
397779
3048
這些都是駕駛者們經常想著的東西
06:56
Now, we can measure測量 this,
141
400827
1663
現在,我們可以量度這些波段
06:58
and we can look at the relative相對的 power功率
142
402490
1985
以及我們可以看看
07:00
between之間 the thetaTHETA waves波浪 and the alphaα waves波浪.
143
404475
2200
theta 波段與alpha 波段的相對強弱
07:02
This gives us a measure測量 of mental心理 workload工作量,
144
406675
2442
這會為我們提供一個大腦活動量的估算
07:05
how much the driver司機 is actually其實 challenged挑戰 cognitively認知
145
409117
3567
看看駕駛者的大腦認知能力
07:08
at any point along沿 the track跟踪.
146
412684
1786
受到多大的挑戰
07:10
Now, we wanted to see if we could actually其實 record記錄 this
147
414470
2942
我們想看看我們能否在賽道上記錄這些數據
07:13
on the track跟踪, so we headed當家 down south to Laguna拉古納 Seca塞卡.
148
417412
3038
因此我們南下到加州的 Laguna Seca 賽車場
07:16
Laguna拉古納 Seca塞卡 is a legendary傳奇的 raceway滾道
149
420450
2016
Laguna Seca 是一個傳奇的賽道
07:18
about halfway between之間 Salinas薩利納斯 and Monterey蒙特雷.
150
422466
2301
它在 Salinas 和 Monterey 兩個縣之間
07:20
It has a curve曲線 there called the Corkscrew螺旋形的.
151
424767
2087
它有一條彎彎的賽道叫做 "開瓶器"
07:22
Now, the Corkscrew螺旋形的 is a chicane強辯, followed其次 by a quick
152
426854
2717
"開瓶器" 有幾個之字形的急彎
07:25
right-handed右手 turn as the road drops滴劑 three stories故事.
153
429571
2746
接著來一個急速的右轉,同時下墜約三層樓的高度
07:28
Now, the strategy戰略 for driving主動 this as explained解釋 to me was,
154
432317
3766
有人告訴我處理這部分賽道的竅門
07:31
you aim目標 for the bush襯套 in the distance距離,
155
436083
2022
你要望著遠處的矮樹
07:34
and as the road falls下降 away, you realize實現 it was actually其實 the top最佳 of a tree.
156
438105
3025
當賽道下降的時候,你會發現那矮樹其實是一棵大樹的頂部
07:37
All right, so thanks謝謝 to the Revs轉速 Program程序 at Stanford斯坦福,
157
441130
3304
好了,因為史丹福大學 Revs 計劃的資助
07:40
we were able能夠 to take John約翰 there
158
444434
1473
我們可以帶約翰到那裏
07:41
and put him behind背後 the wheel
159
445907
964
把他放到一輛1960年生產的
07:42
of a 1960 Porsche保時捷 Abarth阿巴特 Carrera卡雷拉.
160
446871
2439
保時捷Abarth Carrera型號賽車的駕駛員座位上
07:45
Life is way too short for boring無聊 cars汽車.
161
449310
3698
對於開無聊賽車,生命是如此短促!
07:48
So, here you see John約翰 on the track跟踪,
162
453008
1759
你可以看到約翰在賽道上
07:50
he's going up the hill爬坡道 -- Oh! Somebody liked喜歡 that --
163
454767
2184
他正爬上那山丘-- 噢! 有人就是喜歡這樣--
07:52
and you can see, actually其實, his mental心理 workload工作量
164
456951
2465
你可以看到他的腦活動量
07:55
-- measuring測量 here in the red bar酒吧 --
165
459416
2153
-- 紅色那條軸 --
07:57
you can see his actions行動 as he approaches方法.
166
461569
2343
你可以看到他上山時的動作
07:59
Now watch, he has to downshift降檔.
167
463912
3230
現在又看! 他正下坡
08:03
And then he has to turn left.
168
467142
761
接著他作了一個左轉
08:03
Look for the tree, and down.
169
467903
3993
看看那樹,再向下看
08:07
Not surprisingly出奇, you can see this is a pretty漂亮 challenging具有挑戰性的 task任務.
170
471896
2838
可以預料的,你知道這是一個很具挑戰性的任務
08:10
You can see his mental心理 workload工作量 spike as he goes through通過 this,
171
474734
2976
你可以看到他跨越這裏的時候,腦活動量急劇增加
08:13
as you would expect期望 with something that requires要求
172
477710
2064
對於一個這樣複雜的任務
08:15
this level水平 of complexity複雜.
173
479774
2809
我們覺得是合理的
08:18
But what's really interesting有趣 is to look at areas of the track跟踪
174
482583
3416
但有趣的地方是,你可以看到賽道上的某幾點
08:21
where his mental心理 workload工作量 doesn't increase增加.
175
485999
2846
在這些地方約翰的腦活動量沒有增加
08:24
I'm going to take you around now
176
488845
1495
我會帶你往賽道
08:26
to the other side of the track跟踪.
177
490340
1089
的另一端看看
08:27
Turn three. And John's約翰 going to go into that corner
178
491429
2336
第三個彎: 約翰正準備穿越這個轉角
08:29
and the rear end結束 of the car汽車 is going to begin開始 to slide滑動 out.
179
493765
2551
賽車的尾輪開始滑行
08:32
He's going to have to correct正確 for that with steering操舵.
180
496316
2017
他要著手以方向盤改正這個情況
08:34
So watch as John約翰 does this here.
181
498333
2231
看看約翰怎樣做吧
08:36
Watch the mental心理 workload工作量, and watch the steering操舵.
182
500564
2322
看看他的腦活動量,再看看那方向盤
08:38
The car汽車 begins開始 to slide滑動 out, dramatic戲劇性 maneuver演習 to correct正確 it,
183
502886
3672
車子開始滑行,他以戲劇性的操作去改正它
08:42
and no change更改 whatsoever任何 in the mental心理 workload工作量.
184
506558
3523
但腦活動量並沒有改變
08:45
Not a challenging具有挑戰性的 task任務.
185
510081
2832
這並非一個具挑戰性的任務
08:48
In fact事實, entirely完全 reflexive反思.
186
512913
3200
實際上,這完全是個本能反應
08:52
Now, our data數據 processing處理 on this is still preliminary初步,
187
516113
3643
我們在這個情況所得的數據還在最初的階段
08:55
but it really seems似乎 that these phenomenal非凡的 feats功勳
188
519756
2672
但它好像說明了這些驚人的技藝背後
08:58
that the race種族 car汽車 drivers司機 are performing執行
189
522428
1610
賽車手的反應
08:59
are instinctive直覺的.
190
524038
1910
是出於本能的
09:01
They are things that they have simply只是 learned學到了 to do.
191
525948
3390
這些動作都是他們學習的成果
09:05
It requires要求 very little mental心理 workload工作量
192
529338
2282
對於他們要做出這些技藝
09:07
for them to perform演出 these amazing驚人 feats功勳.
193
531620
2824
這些動作對腦活動量的要求並不高
09:10
And their actions行動 are fantastic奇妙.
194
534444
3135
他們的動作實在令人難以置信
09:13
This is exactly究竟 what you want to do on the steering操舵 wheel
195
537579
2611
這些都是每個人希望
09:16
to catch抓住 the car汽車 in this situation情況.
196
540190
3337
可以在這些情況下操作汽車的典範
09:19
Now, this has given特定 us tremendous巨大 insight眼光
197
543527
3445
這為我們提供了寶貴的參考價值
09:22
and inspiration靈感 for our own擁有 autonomous自主性 vehicles汽車.
198
546972
3122
同時亦為我們自己的自動汽車提供相當多的啟發
09:25
We've我們已經 started開始 to ask the question:
199
550094
1928
我們開始問一個問題
09:27
Can we make them a little less algorithmic算法
200
552022
2253
我們能夠將自動汽車發展成較少依賴演算法
09:30
and a little more intuitive直觀的?
201
554275
2449
同時較多依賴直覺的機器嗎?
09:32
Can we take this reflexive反思 action行動
202
556724
2281
我們可以把這樣的本能反應
09:34
that we see from the very best最好 race種族 car汽車 drivers司機,
203
559005
2287
這些我們可以在世界頂級賽車手的身上看到的本能反應
09:37
introduce介紹 it to our cars汽車,
204
561292
1649
結合到我們的汽車上
09:38
and maybe even into a system系統 that could
205
562941
1984
又或者結合到一個系統
09:40
get onto your car汽車 in the future未來?
206
564925
1968
在將來會放在你的汽車上的系統嗎?
09:42
That would take us a long step
207
566893
1611
要將這樣的系統放到汽車上
09:44
along沿 the road to autonomous自主性 vehicles汽車
208
568504
2509
還需要長時間完成的奮鬥
09:46
that drive駕駛 as well as the best最好 humans人類.
209
571013
1912
這個系統將可以與世上最好的車手比擬
09:48
But it's made製作 us think a little bit more deeply as well.
210
572925
3440
但這亦引發出另一個比較深層次的問題
09:52
Do we want something more from our car汽車
211
576365
2968
我們究竟想要一輛能做更多事情的汽車
09:55
than to simply只是 be a chauffeur司機?
212
579333
1840
抑或只是一個司機呢?
09:57
Do we want our car汽車 to perhaps也許 be a partner夥伴, a coach教練,
213
581173
4235
我們是否想汽車變成我們的伙伴、良師
10:01
someone有人 that can use their understanding理解 of the situation情況
214
585408
3087
一個會用它們對環境的理解
10:04
to help us reach達到 our potential潛在?
215
588495
4256
幫助我們發揮潛能的系統呢?
10:08
Can, in fact事實, the technology技術 not simply只是 replace更換 humans人類,
216
592751
2273
究竟,科技是否可以不單單取代人類
10:10
but allow允許 us to reach達到 the level水平 of reflex反射 and intuition直覺
217
595024
4575
而是誘發我們能力所及的
10:15
that we're all capable of?
218
599599
3425
本能反應及直覺呢?
10:18
So, as we move移動 forward前鋒 into this technological技術性 future未來,
219
603024
1923
因此,當我們在科技進步的同時
10:20
I want you to just pause暫停 and think of that for a moment時刻.
220
604947
2821
我希望你停一停,思考這個問題
10:23
What is the ideal理想 balance平衡 of human人的 and machine?
221
607768
3775
甚麼是人類和機器最理想的平衡呢?
10:27
And as we think about that,
222
611543
1709
當我們思考這個問題的時候
10:29
let's take inspiration靈感
223
613252
1731
就讓我們感受一下
10:30
from the absolutely絕對 amazing驚人 capabilities功能
224
614983
3329
人類身體及腦袋
10:34
of the human人的 body身體 and the human人的 mind心神.
225
618312
2816
對我們的啟發吧!
10:37
Thank you.
226
621128
1736
謝謝大家
10:38
(Applause掌聲)
227
622864
4604
(掌聲)
Translated by Jonas Lau
Reviewed by Tien-Hsin Yang

▲Back to top

ABOUT THE SPEAKER
Chris Gerdes - Mechanical engineer
An autonomous car may seem like a thing of the distant future, but mechanical engineer Chris Gerdes is racing to make it a reality today.

Why you should listen

Imagine a car that can drive itself -- that with the push of a button can get you home safely when you’re too tired to drive or have had a night of one too many drinks. Dr. Chris Gerdes , the Director of the Center for Automotive Research at Stanford (conveniently acronymed CARS), and his team are developing a robotic race car, capable of driving at outrageous speeds while avoiding every possible accident. Gerdes’ research focuses on the development of driver assistance systems for collision avoidance, as well as on new combustion processes for engines.

Prior to teaching at Stanford, Gerdes was the project leader for vehicle dynamics at the Vehicle Systems Technology Center of Daimler-Benz Research and Technology North America. His work at Daimler focused on safety analysis.

More profile about the speaker
Chris Gerdes | Speaker | TED.com