TED演讲:一个能防止人群踩踏事件的救生发明(2)（在线收听）

It seemed like a mission impossible, a dream too big, especially for a 15-year-old, yet that dream came true in 2015,

这看起来像不可能完成的任务，完全是异想天开，尤其对于一个15岁的人来讲，但梦想在2015年成真了，

when not only did we succeed in reducing the stampedes and their intensity,

我们不仅成功降低了踩踏事件发生的次数和密度，

but we marked 2015 as the first Nashik Kumbh Mela to have zero stampedes.

也在2015年实现了大壶节历史上首次零踩踏事件的突破。

It was the first time in recorded history that this event passed without any casualties. How did we do it?

这是有历史记录以来第一次这个节日上没有任何人员伤亡。我们是如何做到的？

It all started when I joined an innovation workshop by MIT Media Lab in 2014 called the Kumbhathon

这一切始于我加入的由MIT媒体实验室在2014年设立的名叫大壶马拉松的创新研讨会，

that aimed at solving challenges faced at the grand scale of Kumbh Mela.

目的是解决大壶节面临的大规模人群挑战。

Now, we figured out to solve the stampede problem, we wanted to know only three things:

现在，我们知道要解决踩踏问题，只需要知道三件事情：

the number of people, the location, and the rate of the flow of people per minute.

人数，他们的位置，和每分钟人流的速度。

So we started to look for technologies that would help us get these three things.

所以我们开始寻找可以帮我们得到这三种信息的科技。

Can we distribute radio-frequency tokens to identify people?

我们可以通过无线标记来识别人群吗？

We figured out that it would be too expensive and impractical to distribute 30 million tags.

我们发现，这个方法可能太昂贵且不实用，需要分发超过3千万标记。

Can you use CCTV cameras with image-processing techniques?

能不能用带图像识别的电子摄像头呢？

Again, too expensive for that scale,

同样，对于此等人群规模实在太昂贵了，

along with the disadvantages of being non-portable and being completely useless in the case of rain,

而且还有不够便携的缺点，遇到雨天这样的天气时基本毫无用处，

which is a common thing to happen in Kumbh Mela. Can we use cell phone tower data?

大壶节遇到下雨可是非常平常的事情。我们可以使用蜂窝基站的数据吗？

It sounds like the perfect solution, but the funny part is,

这听起来像是个完美的解决方法，但有趣的是，

most of the people do not carry cell phones in events like Kumbh Mela.

很多人在大壶节这样的节日上，根本不带手机。

Also, the data wouldn't have been granular enough for us.

而且这个数据对我们而言粒度不够。

So we wanted something that was real-time, low-cost, sturdy and waterproof,

所以我们想要真正实时的，低成本的，坚固的，防水的，

and it was easy to get the data for processing.

而且便于获取数据的方案。