Aicha Evans: Your self-driving robotaxi is almost here | TED

翻译人员: Mia Cheng 校对人员: Helen Chang

I’m Aicha Evans,

我是艾莎·埃文斯

I am from Senegal, West Africa,

我来自西非的一个国家，塞内加尔

and I fell in love with technology, science and engineering

在我很小的时候，我就爱上了科技

科学和工程

at a very young age.

Three things happened.

那时候发生了三件事

I was studying in Paris,

那时我在巴黎读书

and starting at seven years old,

从7岁开始上学

flying back and forth between Dakar, Senegal and Paris

在达喀尔、塞内加尔和巴黎 三地之间飞来飞去

as an unaccompanied minor.

是旅途无人陪伴的未成年人

So it wasn't just about the travel.

所以这对我不仅仅是旅途

It was really about a portal to knowledge,

更是让我收获了知识

different environments

应对不同环境时

and adapting.

学会了适应

Second thing that happened

第二件发生的事情是

每次我在塞内加尔的家里

was every time I was at home in Senegal,

I wanted to talk to my friends in Paris.

我都想和我在巴黎的朋友聊天

So my dad got tired of the long-distance bills,

我爸爸不想支付那些长途话费账单

so he put a little lock on the phone --

所以他给电话加了把小锁

the rotary phone.

那种转盘拨号电话

I said, OK, no problem,

我说，好啊，没问题

hacked it,

把锁给破解打开了

and he kept getting the bills.

于是他又收到长途话费账单了

Sorry again, Dad, if you’re watching this someday.

爸爸，如果你某天看到了这个视频 那实在不好意思了

And then, obviously, the internet was also emerging.

后来，显然互联网开始蓬勃发展了

So what really happened was that, in terms of technology,

所以真正发生的事，就科技而言

I really saw it as something that shaped your experiences,

我真的认为它可以改变你的经历

how you understand the world

影响你如何认识这个世界

and wanting to be part of it.

并且想成为它的一部分

And for me,

对于我来说

the common thread is that physical and virtual transportation --

那些经历的共同点就是 实体和虚拟的传输

because that’s really what that rotary phone was for me --

因为当时转盘拨号电话 对我的真正意义

are at the center of the innovation flywheel.

就是创新飞轮的中心

Now, fast-forward.

我们快进到现在

I’m here today,

我今天在这里

I’m part of a movement and an industry

也是作为整个科技行业的一部分

that is working on bringing transportation and technology together.

致力于将交通和科技结合在一起

Huh.

哈

It’s not just about your commutes.

这不仅仅只是影响到你的通勤

It’s really about changing everything

这是在改变一切

改变人们的出行、货物的运输

in terms of how we move people, goods and services, eventually.

最终改变服务业

That transformation involves robotaxis.

这种转变涉及到无人驾驶出租车

Driverless cars again, really?

又是无人驾驶汽车，对吧？

Yeah, yeah, yeah, I’ve heard it before.

对，对，对，我听过这种说法

And by the way, they are always coming the next decade,

顺便说一句，无人驾驶汽车 在未来几十年一定会出现

and oh, by the way,

再顺便说一句

there’s an alphabet soup of companies working on it

有很多顶尖技术的公司都在研究这个

and we can’t even remember who’s who and who’s doing what.

我们甚至记不清，谁是谁，谁在做什么

Yeah?

是吧？

Audience: Yeah.

观众：是

AE: Yeah, OK, well, this is not about personal, self-driving cars.

好的，我要说的不是私人的自动驾驶汽车

Sorry to disappoint you.

如果让你失望了，那抱歉了

This is really about a few things.

我要说的是其他一些事

First of all,

首先

personally and individually owned cars are a wasteful expense,

个人拥有私家车 是一种属于浪费的开支

and they contribute to, basically, a lot of pollution

基本上，只会造成大量污染

and also traffic in urban areas.

以及加重城区里的交通拥堵程度

Second of all, there’s this notion of self-driving shuttles,

其次，我们还有一个概念叫 自动驾驶班车

but frankly, they are optimized for many.

其实就在于服务更多的人群

They can’t take you specifically from point A to point B.

不能专门带你从A点到B点

OK, now we have --

好，我们现在讲了这两个

hm, how am I going to say this --

嗯，我接下来要说的是

the so-called “personal, self-driving” cars of today.

那些今天所谓的私人自动驾驶汽车

Well, the reality is that those cars still require a human behind the wheel.

实际上仍然需要人来驾驶

A safety driver.

一个安全驾驶的司机

Make no mistake about it.

不要误会

I own one of those,

我有一辆这样的车

and when I’m in it,

我在车里

I am a safety driver.

就是那个安全驾驶的司机

So the question now becomes, What do we do with this?

那么现在问题变成了 对此我们该怎么办？

Well, we think that robotaxis,

我们来讲讲无人驾驶出租车

first of all, they will take you specifically from point A to point B.

首先，它们可以专门带你从A点到B点

Second of all, when you're not using them,

其次，当你不用它们的时候

somebody else will be using them.

别人会使用它们

And they are being tested today.

现在无人驾驶出租车已经开始测试了

When I say that we’re on the cusp of finally delivering that vision,

当我说我们即将实现这样的设想

there's actually reason to believe it.

你完全有理由可以相信我

At the core of self-driving technology is computer vision.

自动驾驶技术的核心是计算机视觉

Computer vision is a real-time representation,

计算机视觉就是对环境的实时呈现

digital representation, of the world and the interactions within it.

对于这个世界和其内部交互的数字呈现

It has benefited from leaps and bounds of advancements

计算机视觉技术收益于 现代科技突飞猛进的发展

thanks to computer, sensors, machine learning and software innovation.

尤其是计算机、传感器、机器学习 和软件创新的发展

At the core of computer vision are camera systems.

计算机视觉技术的核心是摄像系统

Cameras basically help you see agents such as cars,

摄像机可以帮你查看周遭环境，像汽车

their locations and their actions,

以及它们的位置和行动

pedestrians,

行人

their locations,

他们的位置

their actions and their gestures.

他们的行为和手势

In addition, there's also been a lot of advancements.

除此之外，还有其他很多进步的地方

So one example is our vehicle can see the skeleton framework

比如，车上可以看到代码框架

to show you the direction of travel;

告诉你行驶的方向

also to give you details, like, are you dealing with a construction worker

还能告诉你其他一些细节信息， 比如，告诉你是否会在施工区域

in a construction zone

遇到建筑工人

or are you dealing with a pedestrian that’s probably distracted

或者你是否碰到了

because they are looking on their phone?

在查看手机而分心的行人

Now the reality, though --

然而现在的实际情况

and this is where it gets interesting --

也是它变得有趣的地方

is that the camera and the algorithms that help us really cannot yet match

就是辅助人类的摄像机和算法

尚无法和人类大脑的能力相提并论

the human brain’s ability to understand and interpret the environment.

无法像人那样理解和解释环境

They just can’t.

就是做不到啊

Even though they provide you really high-resolution imaging

即使它们可以给你提供高分辨率的成像

that really gives you continuous coverage,

真正为你提供连续的实况图像

that doesn’t get fatigued, impaired

不会疲劳，受损

or, you know, drunk or anything like that,

醉酒，或其他类似的情况

at the end of the day,

到头来

there are still things that they can’t see and they can’t measure.

仍然有一些它们看不到 无法测量的情况

So if we want autonomous-driving robotaxis soon,

因此，如果我们想要 尽快实现无人驾驶出租车

we have to supplement cameras.

我们必须要有足够多的摄像机

Let me walk through some examples.

我们来看一些例子

So radar gives you the direction of travel

雷达告诉你行驶的方向

and measures the agent’s movement within centimeters per second.

还测量物体的运动 精确到厘米每秒的范围

Lidar gives you objects and shapes in the real world using depth perception

激光雷达用深度感知 远距离和夜视功能来为你提供

as well as long-range and the all-important night vision.

现实世界中的物体和其形状的信息

And let me tell you about this,

让我再告诉你

because this is important to me personally and people who look like me.

因为这对我个人 和跟我相似的人来说很重要

Then you have, also, long-wave infrared

就是还有长波红外线

where you are able to see agents that are emitting heat,

你可以靠这个看到周围散发热量的物体

such as animals and humans.

像动物和人类

And that’s again,

一样的道理

especially at night,

这功能在晚上

super important.

格外重要

Now, every one of these sensors is very powerful by itself,

每一个传感器各自的功能都非常强大

but when you put them together is when the magic happens.

结合在一起，就是见证魔法的时刻

If you see with this vehicle, for example,

如果你看到这样一辆车

you have these multiple sensor modalities

在车身四个角上

at all top four corners of the vehicle

装有这些不同传感器

that basically provide you a 360-degree field of vision,

它们基本上可以为你提供360度的视野

continuously,

持续不断地

in a redundant manner,

以绝不放过任何细节的态度 来提供信息

so that we don't miss anything.

这样我们就不会错过任何信息

And this is that same thing

其实原理都是一样的

with all of the different outputs fused together.

只是把不同输出的信息汇集在一起

And looking at this, basically,

这样看，基本上

and looking at what we see and how we are able to process the data,

就是收集我们看到的信息数据， 然后看如何处理这些数据

then learn,

再学习

then continue to improve our driving,

然后继续改进自动驾驶的技术

is what tells us that we have confidence,

这让我们更有信心

this is the right approach

这是正确的方法

and this time it’s actually coming.

以及这一次，它真的要实现了

Now, this is not, by the way, a brand new concept, OK?

顺便一提，这不是全新的概念，好吧？

Humans have been basically using vision systems

人类使用视觉系统

to assist them for a long time.

来辅助他们的生活 已经有很长一段时间了

Let me back up the boat a little bit,

我把这个问题退后一步来解释

because I know there’s a question that everybody’s asking,

因为我知道大家都很好奇一个问题

which is, “Hey, how are you going to deal with all the scenarios

就是，如何处理

所有在街上可能会发生的情况？

out there on the streets today?”

Most of us are drivers,

我们大部分人都会驾驶

and it’s complicated out there.

路上的各种交通情况很复杂

Well, the truth is that there will always be edge scenarios

事实上，确实会有很极端的情况

that sit at the boundary of our real-world testing

是我们在真实世界测试很难做到的

or that are just too dangerous to test on real streets.

或者在真实街道上测试太危险了

That is the truth,

这是事实

and it will be the truth for a very long time.

而且这样的事实会持续很长时间

Human beings are pretty underrated in their abilities.

人类严重低估了自己的能力

So what we do is we use simulation.

所以我们要做的就是运用模拟

And with simulation,

通过模拟

we’re able to construct millions of scenarios

我们可以在虚构的环境中

in a fabricated environment

构建出百万个场景

so that we can see how our software would react.

这样就能评估我们的软件反应如何

And that’s the simulation footage.

这就是模拟镜头

You can see we’re building the world,

你可以看到，我们构建虚拟的世界

we’re putting in scenarios

我们设定了各种场景

and we can add things,

我们可以添加一些东西

remove things

也可以拿掉一些东西

and see how we would react.

再看软件会如何反应

In addition, we have what's called a human in the loop.

此外，我们还有所谓的 “有人参与其中“

This is very similar to aviation systems today.

这和当今的航空系统非常相似

We don’t want the vehicle to get stuck,

我们不希望车子陷入无法应对的情况

and there are rare times where it’s not going to know what to do.

极少数情况它不知道该怎么

So we have a team of teleguidance operators

所以我们有一组远程指导操作的人员

that are sitting at a control center,

他们坐在控制中心

and if the vehicle knows that it’s going to be stuck

如果车子知道自己要卡住了

or it doesn’t know what to do,

或者它不知道怎么做

it asks for guidance and help

它可以向指导操作员请求帮助

and it receives it remotely

接着，它远程接受指令

and then it proceeds.

再执行收到的指令

Now, none of these really are new concepts,

这些技术都不是什么新的概念

as I alluded to earlier.

就像我之前说的那样

Vision systems have been assisting humans for a long time,

视觉系统已经辅助人类的生活 很长一段时间了

especially with things that are not visible to the naked eye.

尤其是帮助勘察人类肉眼看不到的东西

So ...

那么

microscopes, right?

像显微镜，对吧？

We’ve been studying microbes and cells for a long time.

我们研究微生物和细胞 已经很长一段时间了

Telescopes:

望远镜

we’ve been studying and detecting galaxies millions of light-years away

我们研究和探测数百万光年之外的星系

for a long time.

也已经很长一段时间了

And both of these have caused us,

而这些都让我们能

for example,

举个例子

to transform industries like medicine,

能改变一些行业，像医药

farming,

农业

astrophysics

天体物理学

and much more.

还有其他更多的行业

So when we talk about computer vision,

所以当我们说计算机系统时

when it started,

当它刚开始发展

it was really a thought experiment

这确实是一场思想的实验

to see if we could replicate what humans see using cameras.

看我们能否用摄像机 复制人类看东西的能力

It has now graduated with sensors,

现在它已经配备了传感器

computers,

计算机

AI

人工智能

and software innovation

还有软件创新

to be about surpassing what humans can see and perceive.

即将超越人类能看到和认知的能力

We’ve made a lot of progress in this field,

我们在这个领域取得了许多进展

but at the end of the day,

但到头来

we have a lot more to do.

我们还有很多事情要做

And with an autonomous robotaxi,

对于无人驾驶出租车

you want it to be safe,

你希望它的每一次出行

right and reliable every single time,

都是安全、正确和可靠的

which requires rigorous testing and optimization.

这需要严格的测试和优化

And when that happens

到那时候

and we reach that state,

我们实现那种状态时

we will wonder how we ever accepted

我们会想知道我们是如何能接受

or tolerated

或容忍

94 percent of crashes

94%的事故

being caused by human [error].

是由于人为（错误）造成的

So with computer vision,

所以有了计算机视觉

we have the opportunity

我们就有机会

to move from problem-solving to problem-preventing.

从解决问题转向预防问题

And I truly, truly believe

我真的相信

that the next generation of scientists and technologists

下一代的科学家和技术人员

in, yes, Silicon Valley,

不仅仅在硅谷

but in Paris,

还要在巴黎

in Senegal, West Africa

西非的塞纳加尔

and all over the world,

以及全世界各地

will be exposed to computer vision applied broadly.

都能够接触到广泛的计算机视觉应用

And with that,

有了计算机视觉的广泛应用

all industries will be transformed,

所有行业都会改变

and we will experience the world in a different way.

我们也将以一种不同的方式 体验这个世界

I hope you can join me in agreeing that this is a gift

我希望你也和我一样觉得

that we almost owe our next generation that is coming,

这是我们欠即将到来的下一代的礼物

because there are a lot of things that computer vision will help us solve.

因为计算机视觉 能帮助我们解决许多问题

Thank you.

谢谢大家

(Applause)

（掌声）