Browse content similar to 15/11/2015. Check below for episodes and series from the same categories and more!
Line | From | To | |
---|---|---|---|
This week, the smart watch yet serious. Virtual reality | 0:00:02 | 0:00:09 | |
skateboarding, and your ceiling lights are talking to us. What do | 0:00:10 | 0:00:15 | |
you mean you can't hear them? -- the smart watch gets serious. | 0:00:16 | 0:00:38 | |
Come on. Hands up if this is you. Your house. Your office. A million | 0:00:39 | 0:00:49 | |
devices, all clamouring for the same Wi-Fi signal. Which means no one | 0:00:50 | 0:00:53 | |
gets a good speed, and the poor old Wi-Fi router becomes Wi-Fry router. | 0:00:54 | 0:01:01 | |
And it's only going to get worse once the Internet of Things properly | 0:01:02 | 0:01:05 | |
gets going and your smartphone has to compete for signal with your | 0:01:06 | 0:01:10 | |
smart shoes, you're smart trousers, your smart mop. That's a good idea. | 0:01:11 | 0:01:15 | |
I'm going to build that. Come with me now to Edinburgh University where | 0:01:16 | 0:01:18 | |
they're giving the Wi-Fi router a break and going online using | 0:01:19 | 0:01:24 | |
something we all plenty of. This laptop is streaming video. It is | 0:01:25 | 0:01:26 | |
connected to the internet but not via Wi-Fi. Via Li-Fi. The ordinary | 0:01:27 | 0:01:37 | |
LED room light is transmitting the data. It is too fast for us to see, | 0:01:38 | 0:01:44 | |
at these LED lights are dimming and writing extremely quickly. And they | 0:01:45 | 0:01:48 | |
are being seen by the laptop as ones and zeros. They are not just | 0:01:49 | 0:01:53 | |
flashing on and off either. There are 256 different levels of | 0:01:54 | 0:01:57 | |
rightness which means eight bits of data can be sent at once. The device | 0:01:58 | 0:02:01 | |
doesn't need to be directly under the light either. It is possible to | 0:02:02 | 0:02:06 | |
detect those flickering bits and bytes from ambient and reflected | 0:02:07 | 0:02:10 | |
light as well. We have seen Li-Fi setups on Click before. Just a few | 0:02:11 | 0:02:16 | |
weeks ago Fujitsu showed off its version in Japan with phones | 0:02:17 | 0:02:20 | |
receiving information from LED strips. But this is simply a 1-way | 0:02:21 | 0:02:26 | |
communication. If you want a proper internet connection, your device | 0:02:27 | 0:02:31 | |
needs to somehow send a signal back as well. And that is what is so new | 0:02:32 | 0:02:36 | |
here in Edinburgh. These laptops not only receive data but send data back | 0:02:37 | 0:02:41 | |
using an infrared uplink. And not just that. As the laptop moves | 0:02:42 | 0:02:46 | |
around the room, it logs onto a different light. That means a room | 0:02:47 | 0:02:53 | |
can serve many devices at once, very much like a mobile phone cellular | 0:02:54 | 0:02:56 | |
network. Time to meet the man who coined the term Li-Fi in his TED | 0:02:57 | 0:03:02 | |
talk four years ago. This is Professor Harold Haas. What are the | 0:03:03 | 0:03:14 | |
advantages of Li-Fi over Wi-Fi? Uses an entirely different part of the | 0:03:15 | 0:03:17 | |
electromagnetic spectrum so it won't interfere with radio or Wi-Fi. | 0:03:18 | 0:03:21 | |
That's why we can add another layer of wireless connectivity. And the | 0:03:22 | 0:03:26 | |
problem with Wi-Fi is that as you increase the number of users, they | 0:03:27 | 0:03:30 | |
will have to split up and divide the bandwidth among the users. And with | 0:03:31 | 0:03:34 | |
Li-Fi, you have another channel there, and you would avoid the | 0:03:35 | 0:03:38 | |
splitting up of bandwidth. And that is going to be really useful when | 0:03:39 | 0:03:43 | |
ever expanding internet of things floods the airwaves with signals. | 0:03:44 | 0:03:46 | |
Your devices will receive data from the ceiling lights, and send data | 0:03:47 | 0:03:53 | |
back possibly by blinking their own tiny power LEDs, like this prototype | 0:03:54 | 0:03:58 | |
here. During the day, at home, I turned my lights off. Does that mean | 0:03:59 | 0:04:03 | |
I don't get data? Certainly you would get data. You can kimberlite | 0:04:04 | 0:04:07 | |
to the level that it would be off for us, for our eyes. But it will | 0:04:08 | 0:04:11 | |
still emit light to provide sufficient data. Does this mean that | 0:04:12 | 0:04:17 | |
we need to make an entirely new type of light, in an entirely new | 0:04:18 | 0:04:22 | |
factory, in order to install this? We can use existing lights to enable | 0:04:23 | 0:04:28 | |
them to be... A Li-Fi transmitter and the thing is we can use the | 0:04:29 | 0:04:33 | |
existing infrastructure. Case in point, this is an off-the-shelf | 0:04:34 | 0:04:37 | |
solar cell which he is being used as a Li-Fi receiver. This light is | 0:04:38 | 0:04:42 | |
transmitting a high def video to the laptop. The great thing about solar | 0:04:43 | 0:04:46 | |
cells as they still work in low light conditions so I can do that in | 0:04:47 | 0:04:51 | |
the video still plays. And that means this kind of receiver will | 0:04:52 | 0:04:54 | |
still work when the devices in your pocket, behind a layer of cotton. | 0:04:55 | 0:05:01 | |
And Harold's team is also working on squeezing more data into the light. | 0:05:02 | 0:05:06 | |
After all, why go white when you've got a whole spectrum to play with? | 0:05:07 | 0:05:12 | |
The next generation of LED lighting will be red, green and blue LEDs and | 0:05:13 | 0:05:17 | |
we mix it in with white as you see here. But the good thing is with the | 0:05:18 | 0:05:21 | |
colours is you can send different data per colour. So red carries a | 0:05:22 | 0:05:25 | |
different stream of information. And I'll show you, if you put in a red | 0:05:26 | 0:05:29 | |
filter here, you see the triangle... The red colour is | 0:05:30 | 0:05:35 | |
transmitting tribals. The red colour is transmitting triangles. And then | 0:05:36 | 0:05:37 | |
you put a blue filter in front of it, you see a rectangle. And we put | 0:05:38 | 0:05:44 | |
in the green filter and that sends in a sawtooth. So three colours, | 0:05:45 | 0:05:49 | |
three different signals, and three times the data. With different | 0:05:50 | 0:05:54 | |
colours we can even go to 100, that is many times faster than we have in | 0:05:55 | 0:06:02 | |
current Wi-Fi. Until something like Li-Fi comes along, one of these is | 0:06:03 | 0:06:06 | |
still taking the strain. And, as most of us know, our Wi-Fi | 0:06:07 | 0:06:10 | |
connection throughout our homes isn't always perfect. So I've got a | 0:06:11 | 0:06:15 | |
few tips, short of changing provider, that could help improve | 0:06:16 | 0:06:20 | |
things. Most people are using a 2.4 gigahertz connection and there are | 0:06:21 | 0:06:26 | |
13 channels. One, six and 11 are the most commonly used. But if you try | 0:06:27 | 0:06:30 | |
an app like this Wi-Fi analyser you can see exactly how many connections | 0:06:31 | 0:06:34 | |
there are. It means that you can then go into your Wi-Fi settings via | 0:06:35 | 0:06:38 | |
your computer and change the channel that you're connecting to, which | 0:06:39 | 0:06:42 | |
should mean less interference from neighbours' Wi-Fi, or even from baby | 0:06:43 | 0:06:47 | |
monitors and microwaves. So first off, it's worth trying a bit of a | 0:06:48 | 0:06:51 | |
cheap and easy DIY solution that I came across. And that's to create | 0:06:52 | 0:06:55 | |
your own reflector. What you'll need is a bit of cardboard, some | 0:06:56 | 0:07:01 | |
tinfoil, and some glue. Stick down the tinfoil, then all you need to do | 0:07:02 | 0:07:05 | |
is folded in a couple of places, and then you place it high on the | 0:07:06 | 0:07:08 | |
router, so that you reflect the signal in the direction that you | 0:07:09 | 0:07:14 | |
want it to go. It may sound a bit unlikely, and look pretty daft, but | 0:07:15 | 0:07:20 | |
in the right situation it can significantly improve your speed. If | 0:07:21 | 0:07:24 | |
you're lucky enough to live somewhere where the size of the | 0:07:25 | 0:07:29 | |
property is an issue, well, Wi-Fi range extender could help. All you | 0:07:30 | 0:07:34 | |
need to do is plug this into a power socket, in an area where it still | 0:07:35 | 0:07:38 | |
going to get connection from your router, and then it should spread | 0:07:39 | 0:07:42 | |
the signal further. Another option is getting a pair of powerline | 0:07:43 | 0:07:46 | |
adapters. One of them will be attached a lead to your router. The | 0:07:47 | 0:07:50 | |
second one will be plugged into any other power socket in the house. You | 0:07:51 | 0:07:54 | |
can then attach your computer with a lead. Many of the new models offer | 0:07:55 | 0:07:59 | |
wireless connection as well, creating a hotspot that smartphones | 0:08:00 | 0:08:03 | |
or tablets can easily connect to. One option is a tri- band router | 0:08:04 | 0:08:08 | |
like this one which as well as having a 2.4 gigahertz connection, | 0:08:09 | 0:08:12 | |
also has two 5 gigahertz connections. That means you may need | 0:08:13 | 0:08:16 | |
to be nearer to the router, but it should provide a stronger, more | 0:08:17 | 0:08:21 | |
consistent connection. Multiple users should get a decent signal at | 0:08:22 | 0:08:26 | |
the same time. And because fewer gadgets use five gigahertz, they | 0:08:27 | 0:08:28 | |
shouldn't be as much interference. Great. The problem is that some of | 0:08:29 | 0:08:34 | |
the older devices you may want to get online with may not use it | 0:08:35 | 0:08:40 | |
either. Not so great. And for some, as prices fall, a high data 4G | 0:08:41 | 0:08:45 | |
contract could be the winner. It could work out cheaper, you won't | 0:08:46 | 0:08:48 | |
even need a home phone connection, and it can have better upload | 0:08:49 | 0:08:53 | |
speeds. What would then come in handy is one of these, a 4G router. | 0:08:54 | 0:08:58 | |
The strength of your connection is of course dependent on the mobile | 0:08:59 | 0:09:02 | |
network you are using. But it does give greater freedom as to where it | 0:09:03 | 0:09:06 | |
can be placed. So play around for long enough and you may find the | 0:09:07 | 0:09:13 | |
perfect spot. It was a difficult week for Facebook this week, as a | 0:09:14 | 0:09:18 | |
Belgian court gave at 48 hours to stop tracking people who are not | 0:09:19 | 0:09:23 | |
members of the social network. Facebook say it will appeal the | 0:09:24 | 0:09:27 | |
decision but could face a fine of up to ?180,000 if it fails to comply | 0:09:28 | 0:09:33 | |
with the court order. And it was the week that SnapChat revealed it got 6 | 0:09:34 | 0:09:37 | |
billion video views every day. Google makes its service available | 0:09:38 | 0:09:41 | |
online and Sony bids fond farewell to Betamax, although it was a | 0:09:42 | 0:09:46 | |
surprise it was making it at all. And a high-speed printer that makes | 0:09:47 | 0:09:49 | |
circuit boards wins and engineering award. Four scientists scooped a | 0:09:50 | 0:09:56 | |
?40,000 prize for their machine. It uses ink to turn circuit board | 0:09:57 | 0:10:00 | |
designs into working prototypes. That is on top of the ?300,000 they | 0:10:01 | 0:10:04 | |
got from kick starter. So clearly they are not doing too badly. And | 0:10:05 | 0:10:09 | |
finally, if you just happen to be in New York this week then you might | 0:10:10 | 0:10:12 | |
believe that a man could fly. This man took a trip around the Big Apple | 0:10:13 | 0:10:19 | |
powered eye gallons of kerosene and two jet engines. Unfortunately it | 0:10:20 | 0:10:22 | |
will only give you about ten minutes of actual flight time before you | 0:10:23 | 0:10:26 | |
have to return to the earth. And before you ask, they have no plans | 0:10:27 | 0:10:31 | |
to sell it to the public. That's cool, I didn't want an awesome jet | 0:10:32 | 0:10:37 | |
pack anyway. Now, this week Apple announced that it has sold over 7 | 0:10:38 | 0:10:41 | |
million of its posh watches. That is more in six months than all other | 0:10:42 | 0:10:46 | |
smart watches combined, in the last year. Goodness! At there is one | 0:10:47 | 0:10:51 | |
thing we think all smart watches still share, in our humble opinion. | 0:10:52 | 0:10:56 | |
They look really, really, really, really geeky. Now, though, it is the | 0:10:57 | 0:11:03 | |
return of the classic watch brands to try and make something special. | 0:11:04 | 0:11:09 | |
We sent Dave Lee to New York. Thanks to them, the Swiss watch industry is | 0:11:10 | 0:11:19 | |
connected. Today is the marriage of watch Valley and Silicon Valley. We | 0:11:20 | 0:11:28 | |
are giving birth to eternity in a box -- thanks to TAGHeuer. We have a | 0:11:29 | 0:11:37 | |
computer inside the watch. Now, in case he didn't quite get that, what | 0:11:38 | 0:11:42 | |
Jean-Claude was trying to say was that his company, TAGHeuer, is | 0:11:43 | 0:11:49 | |
making its first ever smart watch. TAGHeuer has more than 150 years of | 0:11:50 | 0:11:53 | |
watchmaking in the bag, and this watch is about saying to Apple, the | 0:11:54 | 0:11:58 | |
Samsung, to Motorola, all those other companies, they just don't | 0:11:59 | 0:12:00 | |
know what they're doing when it comes to making a stylish timepiece. | 0:12:01 | 0:12:05 | |
But will this be enough to turn the opinions of people who just love | 0:12:06 | 0:12:09 | |
classic watches? Yes. Certain people, yes. Not everybody, of | 0:12:10 | 0:12:15 | |
course. Certain people, it is good. You are wearing two watches at the | 0:12:16 | 0:12:19 | |
moment. Yes. One of them is a smart watch, the other one is on. Yes, and | 0:12:20 | 0:12:25 | |
you hardly see the difference. I'm going to wager you will take the | 0:12:26 | 0:12:29 | |
smart watch fairly certain. Not necessary. Why? Because it is very | 0:12:30 | 0:12:36 | |
light, it is very comfortable, and then it's quite nice to play with | 0:12:37 | 0:12:41 | |
it! The watch is powered by Google's Android wear. Meaning it | 0:12:42 | 0:12:47 | |
has all the existing apps made for other watches until now. And it has | 0:12:48 | 0:12:52 | |
that authentic TAGHeuer look. On the inside, the electronics are made by | 0:12:53 | 0:12:57 | |
Intel, a company which is desperate to make sure it doesn't miss out on | 0:12:58 | 0:13:00 | |
the smart watch industry in the same way it was left behind with | 0:13:01 | 0:13:05 | |
smartphones. This is a part of us not being absent from the next big | 0:13:06 | 0:13:09 | |
thing. And I think, you know, you're seeing our efforts in wearable | 0:13:10 | 0:13:14 | |
devices. You saw us make some acquisitions this year in wearable | 0:13:15 | 0:13:18 | |
devices. Both for the wrist and for the head worn eyeglasses, and there | 0:13:19 | 0:13:23 | |
are many other segments of devices that we are making sure that we | 0:13:24 | 0:13:28 | |
don't miss the next big thing. But unfortunately the watch still | 0:13:29 | 0:13:32 | |
suffers from the same problems as many of its competitors, the battery | 0:13:33 | 0:13:37 | |
just about lasts a day, and it's a it big. So the watch is quite big, | 0:13:38 | 0:13:42 | |
quite chunky, safe to say it is definitely a man's watch. Are you | 0:13:43 | 0:13:48 | |
hoping to make a Lady's smart watch as well? We need a lady's watch. A | 0:13:49 | 0:13:55 | |
woman, you know, the phone rings, will she find it in her bag? The | 0:13:56 | 0:14:02 | |
phone has stopped ringing. So for women, it is quite practical to have | 0:14:03 | 0:14:05 | |
information on the wrist. But it will never compete with an elegant, | 0:14:06 | 0:14:09 | |
wonderful watch, for sure. This will set you back a tasty | 0:14:10 | 0:14:22 | |
$1500. It is not a perfect device but they seem pretty confident. | 0:14:23 | 0:14:26 | |
Although we get the feeling he might always be like this. LAUGHTER. Well, | 0:14:27 | 0:14:33 | |
there is a man who loves his job. I wonder what he would make of this, | 0:14:34 | 0:14:38 | |
possibly the most unusual smartwatch I have seen in a while. This does | 0:14:39 | 0:14:44 | |
not just tell the time calls and messages, it knows what you want. | 0:14:45 | 0:14:51 | |
All these objects you interact with, your phone, your tablet, even | 0:14:52 | 0:14:57 | |
your refrigerator, they admit -- emit electronic noise. And that | 0:14:58 | 0:15:02 | |
flows into the body because the body is conductive. Once that signal goes | 0:15:03 | 0:15:06 | |
into the body, the smartwatch since on your wrist can sense that signal. | 0:15:07 | 0:15:10 | |
And these signals are very characteristic. Each object has its | 0:15:11 | 0:15:16 | |
own little magnetic profile. OK. Applications like what? Researchers | 0:15:17 | 0:15:23 | |
say it can unlock your computer or give you real time directions simply | 0:15:24 | 0:15:32 | |
by touching an appropriate object. So this could really be a properly | 0:15:33 | 0:15:35 | |
smartwatch, one that will know when you are rushing your teeth and start | 0:15:36 | 0:15:40 | |
timing you, or one that knows you are about to lose and drilling and | 0:15:41 | 0:15:43 | |
brings up an instruction manual on how to do it. But a really, really | 0:15:44 | 0:15:49 | |
smart advice -- device might advise you to read that first before... | 0:15:50 | 0:15:57 | |
Well, maybe that is just me. One of the problems of virtual reality we | 0:15:58 | 0:16:00 | |
have talked a lot about recently is the fact you can see and hear your | 0:16:01 | 0:16:03 | |
surroundings but you cannot feel what is going on and that rather | 0:16:04 | 0:16:06 | |
breaks the illusion. For example, wouldn't it be great if you were | 0:16:07 | 0:16:10 | |
virtual reality skate boarding to be able to feel the ground beneath your | 0:16:11 | 0:16:13 | |
feet. That has been the question on my lips for months now, I promise. | 0:16:14 | 0:16:17 | |
Fortunately, we have found a solution in Japan. Who doesn't want | 0:16:18 | 0:16:27 | |
a vibrating virtual reality skate boarding simulator? Yes, the Tokyo | 0:16:28 | 0:16:38 | |
Institute of Technology has built this amazingly involved boardgame. | 0:16:39 | 0:16:44 | |
This is brilliant. These skateboarders vibrating... Can you | 0:16:45 | 0:16:53 | |
hear the difference? The cobbles! It lets riders feel the ground texture | 0:16:54 | 0:16:56 | |
change beneath their feet as if they were skateboarding on real ground. | 0:16:57 | 0:17:02 | |
And this is the best bit. Because I'm in virtual reality, I can get | 0:17:03 | 0:17:05 | |
all of the thrills of high speed riding in real life with none of the | 0:17:06 | 0:17:10 | |
consequences when I fall off or inevitably bash into the scenery. | 0:17:11 | 0:17:16 | |
Normally when you were virtual reality goggles, you can't | 0:17:17 | 0:17:26 | |
experience things in the same way. I'm going to stop for a second | 0:17:27 | 0:17:28 | |
because I'm meant to be telling you about this. That is really | 0:17:29 | 0:17:37 | |
incredible! Adding that sensory element really does bring a game | 0:17:38 | 0:17:40 | |
like this to life. The sound, the feeling of the vibration and | 0:17:41 | 0:17:44 | |
certainly the actual physical peril. I really was quite reliant on | 0:17:45 | 0:17:50 | |
this hole and I did not think I would need that at all. To get a | 0:17:51 | 0:17:57 | |
sound, they mounted microphones underneath a real skateboard and | 0:17:58 | 0:17:59 | |
record the sound of different road textures. Then the sound for it | 0:18:00 | 0:18:04 | |
texture is played back during the game depending on the texture being | 0:18:05 | 0:18:09 | |
played -- displayed on the screen. Underneath, we have a pair of 50 | 0:18:10 | 0:18:13 | |
watts bass speakers. Basically, it is using low-frequency sound to give | 0:18:14 | 0:18:19 | |
that sense of vibration. And then under here, we have a collection of | 0:18:20 | 0:18:22 | |
microcontrollers, of are sending data back to the Peter. You have -- | 0:18:23 | 0:18:30 | |
back to the computer. You have your skateboard, your treadmill, your | 0:18:31 | 0:18:35 | |
amplifier and the oculus rift. It is a very ambitious project but it does | 0:18:36 | 0:18:42 | |
work. The system also recognises when you tilt the skateboard. The | 0:18:43 | 0:18:44 | |
computer-generated world responds accordingly. Plus there is a sense | 0:18:45 | 0:18:51 | |
or in the treadmill. And when you kick the treadmill, it detects the | 0:18:52 | 0:18:57 | |
force. And the beauty matches speed to the force. -- the computer. And | 0:18:58 | 0:19:06 | |
no, it is not as easy as it looks. Visit our producer trying it for the | 0:19:07 | 0:19:12 | |
first time. -- this is our producer. LAUGHTER. Frivolity aside, | 0:19:13 | 0:19:24 | |
I was impressed by how easily I accept it and learn from feedback in | 0:19:25 | 0:19:28 | |
the virtual world. This is massive intentional. A world that feels real | 0:19:29 | 0:19:34 | |
enough to landscape or in, no matter what level of reality you are | 0:19:35 | 0:19:38 | |
actually in. -- real enough to learn to skateboard in. And from one | 0:19:39 | 0:19:45 | |
exhilarating ride to another. The London Underground is a busy bit of | 0:19:46 | 0:19:51 | |
old bump and grind stone. Sun plans to visit. Even without an incident, | 0:19:52 | 0:19:55 | |
some parts get clogged just by people waiting to get home. And when | 0:19:56 | 0:20:02 | |
things do start bundling up, this is where the decisions are made about | 0:20:03 | 0:20:07 | |
what to do. I love the fact that we have all of these big bits that are | 0:20:08 | 0:20:15 | |
not on the Tube map. You can do such clever things with trains. The | 0:20:16 | 0:20:18 | |
oldest Tube network in the world spans over 1000 kilometres and if | 0:20:19 | 0:20:22 | |
something goes wrong in one section, it can affect other | 0:20:23 | 0:20:25 | |
sections hours later. The team in this control room try to keep things | 0:20:26 | 0:20:30 | |
moving to minimise delays. All trains report their whereabouts and | 0:20:31 | 0:20:34 | |
over the last decade or so, some have been upgraded to the fully | 0:20:35 | 0:20:38 | |
automated, the driver here becoming more of a supervisor. The strains | 0:20:39 | 0:20:44 | |
automatically change speed to maintain the gaps between them and | 0:20:45 | 0:20:47 | |
regulate the service. -- these trains. Some have been fitted with | 0:20:48 | 0:20:53 | |
scales, weighing the load and estimating the number of passengers | 0:20:54 | 0:20:57 | |
on board. Even with all this information, the plans in place to | 0:20:58 | 0:21:00 | |
avoid overcrowding and delays are surprisingly low-tech. At the moment | 0:21:01 | 0:21:06 | |
for the Central line, we would look at the pinch points that we have. A | 0:21:07 | 0:21:11 | |
team appear would be bringing around different locations on the network | 0:21:12 | 0:21:14 | |
where we know we have high volumes of cars is and where experience | 0:21:15 | 0:21:19 | |
tells us we have pinch points. -- high volumes of customers. And based | 0:21:20 | 0:21:26 | |
on the way the network operates at the moment, we have additional areas | 0:21:27 | 0:21:31 | |
of the network where other lines will take the normal loading for | 0:21:32 | 0:21:35 | |
those sections. We also have a lot of contingency plans, well versed | 0:21:36 | 0:21:40 | |
plans and our teams then go into to address these kinds of issues. If we | 0:21:41 | 0:21:46 | |
have crowding issues or potential capacity issues around Oxford | 0:21:47 | 0:21:51 | |
Circus, we have a team there is well-trained and that practices a | 0:21:52 | 0:21:54 | |
lot and they will those plans to help us. Will helm has developed an | 0:21:55 | 0:22:02 | |
algorithm that can predict the knock-on problems across a network | 0:22:03 | 0:22:06 | |
over the next couple of hours. Compressed two years of work into | 0:22:07 | 0:22:11 | |
three sentences, please. We take the information out of the train system, | 0:22:12 | 0:22:15 | |
where the train is at a certain time, and we look at how that looks | 0:22:16 | 0:22:19 | |
in a pattern of historical data. And then we say that this is a pattern | 0:22:20 | 0:22:25 | |
we have seen before. And then we can predict. The system sucks up all of | 0:22:26 | 0:22:29 | |
the real-time data from the Stockholm rail network, not just | 0:22:30 | 0:22:33 | |
will helm 's very lovely train set here, and repeatedly makes | 0:22:34 | 0:22:37 | |
predictions every single minute. That means as an incident develops | 0:22:38 | 0:22:42 | |
or resolves, the numbers change. What will the officials be able to | 0:22:43 | 0:22:47 | |
do with this information? If they see there is a problem and this is | 0:22:48 | 0:22:49 | |
how it will look into hours, what can they do? What they want to do is | 0:22:50 | 0:22:56 | |
screw my model up. You say there will be delayed, they say they are | 0:22:57 | 0:22:59 | |
going to solve that. The problem today is that we are looking at | 0:23:00 | 0:23:04 | |
trains, not humans. If you combine that with information as to how has | 0:23:05 | 0:23:08 | |
the train is, you can say that you have a very heavy train, lots of | 0:23:09 | 0:23:14 | |
people on board, and then say let us run by train all the way through. | 0:23:15 | 0:23:17 | |
That will minimise the total amount of delays for humans. But today, all | 0:23:18 | 0:23:24 | |
of those systems look at the train. The ultimate goal is to get these | 0:23:25 | 0:23:28 | |
numbers is close to the real arrival time as possible and, of course, no | 0:23:29 | 0:23:33 | |
more last-minute cancellations. In fact, travellers and dock when will | 0:23:34 | 0:23:41 | |
still have access to -- soon have access to the information themselves | 0:23:42 | 0:23:45 | |
via the app. Incidentally, do you know which underground station the | 0:23:46 | 0:23:48 | |
guys at transport for London told me they had to close most often because | 0:23:49 | 0:23:55 | |
of overcrowding? Victoria. Correct. Six o'clock every evening, this | 0:23:56 | 0:24:02 | |
place is crazy. I wonder if anyone would mind if I knocked off early | 0:24:03 | 0:24:07 | |
tonight and beat the rush. See | 0:24:08 | 0:24:08 |