High Tech Highway Click


High Tech Highway

Click visits the US to see how predictive analytics help in emergencies. Plus Honda's work on creating robots with emotions and some AR pilot training.


Similar Content

Browse content similar to High Tech Highway. Check below for episodes and series from the same categories and more!

Transcript


LineFromTo

That's it from me.

0:00:000:00:01

Now on BBC News, Click.

0:00:010:00:06

This week, cars, bars and a police

riot. -- ride.

0:00:090:00:23

I am on my way to a reported

incident on 1-off Las Vegas's

0:00:400:00:45

busiest highways. With the last rain

falling over four months ago, the

0:00:450:00:53

Audley Road is mixed with fresh

water have become a lethal recipe

0:00:530:00:57

for disaster. -- oily roads. In the

driving seat is a sergeant from the

0:00:570:01:04

Nevada Highway Patrol. He is using a

software that alerts into an

0:01:040:01:08

incident as soon as it is reported

by someone calling 911 or through

0:01:080:01:15

driving apps and provide him with

details and the best route to get to

0:01:150:01:19

the scene.

The location, what kind

of accident, the degree, and if

0:01:190:01:25

there are any responders that are on

their way.

It constantly updates in

0:01:250:01:31

on the situation as it develops.

Having a robust system in place

0:01:310:01:35

doesn't just help with

weather-related collisions.

With our

0:01:350:01:41

Route 91 shooting that we had, for

the portion that we handled, the

0:01:410:01:49

Highway Patrol, it's getting the

public off the highway as quickly as

0:01:490:01:53

possible, closing of the freeways we

could have the critical resources,

0:01:530:01:57

fire, medical, ambulances, to get

people to the hospital quickly.

In

0:01:570:02:03

2017 15,000 crashes were attended

to, with over 300 people dying on

0:02:030:02:08

average in each year in Rosa -- road

accidents in Nevada, getting

0:02:080:02:13

emergency services to be seen as

quickly as possible is critical.

0:02:130:02:16

There is an injury. Camera 217.

The

system has been running through the

0:02:160:02:23

regional transportation commission's

traffic centre for the past three

0:02:230:02:28

months.

Now because we are getting

information through so many

0:02:280:02:31

different data streams, not just

dispatches, but social media, things

0:02:310:02:37

like the apps. Because all of this

is happening so quickly we might

0:02:370:02:42

have already sent out all of that

information and had everybody in

0:02:420:02:45

this room where before the first 911

call comes in. So we are talking

0:02:450:02:50

about possibly ten to 15 minutes of

improvement in response time in some

0:02:500:02:56

of these incidents. That's major

when you are dealing with these

0:02:560:03:00

kinds of incidents.

The app pulls

information from various sources, in

0:03:000:03:06

vehicle sensors, TV cameras,

information from driving apps. It

0:03:060:03:10

factors in what day of the year it

is, the time of day and the weather.

0:03:100:03:15

Responding to incidents rapidly is

one thing, but the point is to be

0:03:150:03:19

able to predict them before they

happen so the responders can be

0:03:190:03:24

better prepared and in the right

location.

We look at the historical

0:03:240:03:29

data, running through algorithms to

develop patterns that are merging.

0:03:290:03:34

By doing that we can look forward in

time to identify where these

0:03:340:03:40

incidents are likely to occur.

Unfortunately, the app wasn't able

0:03:400:03:52

to predict this one.

It looks like

it is the real. You can see how she

0:03:520:03:57

was spinning out. She did a full 180

and struck right here.

Being able to

0:03:570:04:03

foresee accidents here could really

save lives. The hope is that as the

0:04:030:04:09

data gets more sophisticated the

predictions will become more

0:04:090:04:12

accurate.

Everyday we get more and

more evidence about what causes,

0:04:120:04:17

what triggers, incident and the

artificial learning get smarter and

0:04:170:04:22

smarter and more capable.

For Nevada

now the initial results are

0:04:220:04:26

promising.

They get there faster, we

clear it faster and that means less

0:04:260:04:32

secondary accidents and if you think

about it secondary accidents have

0:04:320:04:38

basically 18% of secondary accidents

are fatalities. So we are reducing

0:04:380:04:43

the fatalities on the road.

And of

course the goal is to prevent

0:04:430:04:47

accidents altogether and Richard

Taylor and Lara Lewington have been

0:04:470:04:51

looking at some in car technologies

that may help make that a reality.

0:04:510:05:00

At CES as you might expect there's a

lot of interest in self driving cars

0:05:000:05:04

and it's pretty clear that we are on

a 1-way street towards full

0:05:040:05:08

autonomy.

But that does still seem

to be a way off, although we don't

0:05:080:05:12

know exactly how far. In the

meantime there is plenty of

0:05:120:05:15

innovation to be seen before we

reach our final destination.

0:05:150:05:21

Unsurprisingly, the move towards

autonomy to driving is focused

0:05:210:05:25

largely on safety, with healing day

creating a system to intervene when

0:05:250:05:30

we needed the most. -- with Hyundai.

With accommodation of biometric

0:05:300:05:35

sensors in the seat, they are

tracking heart rate and a low

0:05:350:05:38

resolution camera which is tracking

your facial movements. The reason it

0:05:380:05:42

is low resolution is so that the

refresh rate is quicker. So if

0:05:420:05:46

there's a problem, if you've lost

concentration or you are drifting

0:05:460:05:49

off to sleep, then the car can

quickly react. So autonomously be

0:05:490:05:53

moved off the road to a safe spot.

And the basic tremors of this

0:05:530:05:58

technology could be available in

just a year. -- a sick premise.

0:05:580:06:05

Meanwhile, Nissan has a more

futuristic twist on biometrics. The

0:06:050:06:09

idea of this system is really to

provide an interaction between man

0:06:090:06:14

and machine, between my brain and

the AI. The concept here with Nissan

0:06:140:06:21

is even in a world of autonomous

vehicles, there will be roles for

0:06:210:06:26

humans to play. After all, a lot of

people find driving a positive

0:06:260:06:31

experience. It can interpret signals

coming from the human and actually

0:06:310:06:36

in -- enhance the ride. This sort of

brain the vehicle text currently

0:06:360:06:42

involves wearing this bizarre helmet

to capture my brain activity and

0:06:420:06:46

interpret signals as much as half a

second before my muscles do. So, as

0:06:460:06:51

I'm about to say change lane or hit

the brakes on it will initiate the

0:06:510:06:55

action for me, giving me a smoother

ride, and yet still allowing me a

0:06:550:06:59

sense of control. They do need to

sort out that helmet, though. Oh

0:06:590:07:03

dear. I'm not driving very well

here.

Yet what we can't hide away

0:07:030:07:08

from is the fact that when full

autonomy does come to pass, it is

0:07:080:07:13

not simply about cars. This is

Yamaha's concept motorbike. A self

0:07:130:07:19

driving racing vehicle that should

be able to do speeds of over 120

0:07:190:07:24

mph, although not on actual roads

you would hope. But whatever the

0:07:240:07:29

form of autonomous vehicle it will

need to interact safely with

0:07:290:07:33

pedestrians at and cyclists, a

challenge that Ford are hoping to

0:07:330:07:39

overcome in their vehicles.

Initially cyclists will have to be

0:07:390:07:42

seen by the vehicles and we are

building the reception into our

0:07:420:07:45

autonomous vehicle that allows it to

detect the cyclists, objects,

0:07:450:07:50

understand their intent and make

sure we can be safely navigating

0:07:500:07:53

same space.

And Ford are just one of

the big brands that have called on

0:07:530:07:59

the help of a company, whose

processes, combined with

0:07:590:08:03

intelligence software, can make the

environment around the vehicles

0:08:030:08:06

safer. For example using light as

sensors to alert the driver who is

0:08:060:08:10

about to open a car door onto a

cyclist. -- lidar sensors.

And AI is

0:08:100:08:16

fuelling other experiences. Speech

recognition specialists power many

0:08:160:08:20

of the day's in car interactions how

they are looking in the future as

0:08:200:08:25

well.

Today we think about

interacting with using voice, but

0:08:250:08:32

there are other modalities. Of

course we have a touchscreen, but

0:08:320:08:36

maybe we can use gestures and in

this particular prototype we

0:08:360:08:43

introduced eye tracking, to help the

assistant understand what am I as a

0:08:430:08:48

driver looking at and then I can ask

questions about my environment. If I

0:08:480:08:52

see a coffeeshop in front of me, I

can just ask a question about it.

0:08:520:08:57

What is the user rating of this copy

shop?

Starbucks coffee has a user

0:08:570:09:03

rating of three stars.

So the other

part of the system is that there are

0:09:030:09:08

microphones in different parts of

the car, which means the AIA can

0:09:080:09:11

respond according to whether

different passengers are. So here on

0:09:110:09:15

the passenger seat I can say,

"hello, Dragon, I'm cold".

OK,

0:09:150:09:20

raising the temperature in zone

two...

0:09:200:09:25

There's definitely a trend towards

making our journeys more enjoyable

0:09:250:09:29

as well as safer. To you to have

updated their happiness concept,

0:09:290:09:35

aiming for a more pleasurable

journey and even suggesting where

0:09:350:09:38

you might want to go -- Toyota. For

anyone needs their car to tell them.

0:09:380:09:43

I will tell you something

interesting. There are many options

0:09:430:09:49

around Union Square for casual

dining the Michelin stars. Do you

0:09:490:09:53

like it?

Yes. That was a bit of fun,

but I didn't need the AI to tell me

0:09:530:10:02

that I was ready for dinner. Let's

go.

0:10:020:10:04

Welcome to the Week in Tech. The

week Ford announced it would have £8

0:10:100:10:15

billion in electric cars in the next

five years. A flaw in a VR app left

0:10:150:10:20

20,000 users' names exposed. And

hackers stole $400,000 worth of

0:10:200:10:27

cryptic currency by hijacking a

server. It was a busy week for

0:10:270:10:32

crypto currency, as big coin

encountered its busiest daily crash

0:10:320:10:36

in four months. -- biggest. It is

thought fears over increased

0:10:360:10:41

regulation, especially in Asia,

would be an issue. A contraceptive

0:10:410:10:45

app previously thought as effective

as the pill has been criticised by a

0:10:450:10:49

Swedish hospital for a number of

unintended pregnancies they say were

0:10:490:10:52

linked to the app. The company

behind it have defended the product,

0:10:520:10:56

saying that as with any form of

contraception it isn't 100%

0:10:560:11:00

effective. They are now launching an

internal investigation, however. And

0:11:000:11:04

I bet you didn't expect the latest

Nintendo offering to include a whole

0:11:040:11:08

lot of cardboard. The latest add-ons

for the switch console are cardboard

0:11:080:11:13

packs, turning the consoles into a

fishing rod, motorbike and even a

0:11:130:11:18

robot suit. Gimmick or brilliant?

Finally, the rescue with a

0:11:180:11:22

difference. A drone was used to save

two swimmers off the coast of NSW in

0:11:220:11:27

Australia. Lifeguards were being

trained to use the rescue drone when

0:11:270:11:31

practice became reality and it was

launched, robbing a flotation device

0:11:310:11:35

to the teenagers. The whole rescue

took just 72nd. -- 70 seconds.

0:11:350:11:44

In this trendy part of downtown Las

Vegas, these passengers are waiting

0:11:440:11:48

to hop on a special kind arrive. For

the past two months, this French

0:11:480:11:54

autonomous vehicle company has been

offering free bus ride to the

0:11:540:11:58

public. Admittedly it doesn't travel

far, it just does a loop around the

0:11:580:12:02

block with one stop at a doughnut

shop. At least they are getting a

0:12:020:12:07

taste of the future! Down the road I

am waiting to catch a more private

0:12:070:12:13

road which I've built on an app --

booked. As if by magic the door

0:12:130:12:20

opened! The team was still ironing

out a few issues, shall we say. I

0:12:200:12:28

think this is the first genuinely

autonomous vehicle I've been in

0:12:280:12:35

where there really is no driver and

their really is no place for a

0:12:350:12:40

drive. There's just a safety man

here. That's it. Safety man has an

0:12:400:12:49

Xbox one controller down by his

side. NAVYA is not alone in this

0:12:490:12:55

space. Other companies have been

battling it out to become the first

0:12:550:12:59

fully autonomous cab sharing

service. Self driving cars use a lot

0:12:590:13:05

of sensors to be able to navigate

the road safely. That's one of the

0:13:050:13:08

most important is LIdar, how the car

judges its surroundings. The design

0:13:080:13:16

of these centres is at the heart of

a court case. NAVYA's car is no

0:13:160:13:22

different. It also uses Lidar to

look around. What it is not doing is

0:13:220:13:28

looking at the traffic lights to

judge what colour they are. They've

0:13:280:13:32

fitted special sensors to each

traffic light and those sensors talk

0:13:320:13:35

to the car. That doesn't sound very

scalable to me. That sounds like you

0:13:350:13:40

wouldn't be able to put this sort of

technology on the open road without

0:13:400:13:43

fitting every single traffic light

in the US with these centres. It is

0:13:430:13:47

much more just for predetermined

routes for this kind of shuttle

0:13:470:13:51

vehicles.

While I've been riding around in

0:13:510:13:58

this particular smart vehicle, Dave

Lee has been up in Reno, not that

0:13:580:14:02

far away, looking at a system that

is making use of data collected by

0:14:020:14:07

vehicles like this to help an entire

city to move more smoothly.

0:14:070:14:11

There's been great strides made in

self driving technology over the

0:14:190:14:22

past decade or so, but the thing

about autonomy is that it often

0:14:220:14:26

tested in bright and clear

conditions. The real world is much

0:14:260:14:30

more distracting.

0:14:300:14:36

conditions. The real world is much

more distracting. In fact, it is not

0:14:360:14:38

just

0:14:380:14:38

more distracting. In fact, it is not

just darkness that is difficult for

0:14:380:14:41

existing autonomous technologies.

Whether it is through rain, snow, or

0:14:410:14:45

just Faarup ahead on the road, there

is a lot self driving vehicles

0:14:450:14:50

struggle to see. Important work is

taking place at the University of

0:14:500:14:55

Reno, Nevada, that is attempting to

solve that problem, making autonomy

0:14:550:15:00

more intelligent. And it all begins

here, on the corner of 15th and

0:15:000:15:07

Virginia.

So at that corner we have

a light sensor. That light sensor

0:15:070:15:12

used to be on the autonomous

vehicle. But if we move it from the

0:15:120:15:16

vehicle to the intersection, so it

can track each pedestrian here, each

0:15:160:15:20

vehicle here.

What kind of things is

that picking up? Is it recognising

0:15:200:15:25

who people are?

No, it only

recognises this as a pedestrian or

0:15:250:15:29

this is a vehicle. It does not

recognise who the person is.

Think

0:15:290:15:34

of this intersection as providing

more eyes to an autonomous vehicle.

0:15:340:15:39

It could detect the threat and

communicate that to a car heading in

0:15:390:15:43

its direction, telling it to slow

down, beware.

So what these centres

0:15:430:15:48

are doing in essence is giving

autonomous car is more eyes on the

0:15:480:15:52

road. Yes.

They just know more about

what is coming up ahead.

Exactly, so

0:15:520:15:57

no black spots.

Part of the same

programme is this connected car. A

0:15:570:16:06

modified Lincoln that can not only

drive itself around, but also

0:16:060:16:10

communicate with other vehicles and

components in the city, signalling

0:16:100:16:14

its intention is to others.

The

hardware that you see is pretty

0:16:140:16:18

similar to what you are going to see

in most autonomous vehicles, if not

0:16:180:16:22

all of them. Where we really dissing

wish ourselves as in the software.

0:16:220:16:26

So our research focuses on what I

call social intelligence. We are

0:16:260:16:30

trying to build machines that

understand people, and understand

0:16:300:16:34

human social behaviour, and can

predict what other people are going

0:16:340:16:38

to do, and then act appropriately.

It is a skill that humans have. We

0:16:380:16:42

navigate driving effortlessly, even

though we can't read other people's

0:16:420:16:46

mines, and it is a skill that

computers are going to have to have

0:16:460:16:50

if they are ever going to drive cars

in the world with the rest of us.

0:16:500:16:54

And then there is the challenge of

making the technology work in

0:16:540:16:57

difficult conditions. Inspired by an

earlier project to help drone see in

0:16:570:17:01

the dark, the team at the

University's autonomous robots lab

0:17:010:17:04

has confined lidar, radar and

cameras, to dramatically improve

0:17:040:17:12

what car can copperhead. It is also

cheap. Once that technology is safe

0:17:120:17:19

and ready, the plan is to deploy it

on electric losses like this one.

0:17:190:17:23

Until then, the team plans to use

the autonomous tech together large

0:17:230:17:27

amounts of data in preparation for a

self driving future. This bus made

0:17:270:17:34

by a California -based company is

already out on Reno's roads, but

0:17:340:17:38

right now with a more traditional

type of driver.

It is not autonomous

0:17:380:17:43

yet. The idea is to at some point

focus on that project. However,

0:17:430:17:47

right now we are focusing on data

collection for what we call the

0:17:470:17:51

living lab, and data collection is

going to be used for the mobility

0:17:510:17:56

programme.

For the foreseeable

future, these buses will gather data

0:17:560:17:59

for the living lab programme in

Reno, a city that perhaps knows more

0:17:590:18:04

about what is going on on its

streets than almost any other city

0:18:040:18:08

in the world. That was Dave, and now

the something we have been hearing a

0:18:080:18:17

lot about recently. Augmented

reality. Now, it works by overlaying

0:18:170:18:21

graphics on top of the real world,

and while AR games like Pokemon Go

0:18:210:18:27

have enjoyed global success, the

most hyped it of AR kit, Magic to

0:18:270:18:32

get leap, is still waiting to be

released. AR remains a technology

0:18:320:18:36

that promises more than it delivers

-- Magic Leap. But by combining AR

0:18:360:18:42

with AI, researchers in Florida are

hoping to create new ways to train

0:18:420:18:46

people to perform complex tasks. We

took them AR kit for a test drive,

0:18:460:18:52

or should that be a test flight? The

University of Central Florida has a

0:18:520:18:59

long established relationship with

the simulation industry. Helping

0:18:590:19:04

create simulated experiences for

everything from driving to

0:19:040:19:08

supermarket shopping. The simulation

lab here's latest project is a bit

0:19:080:19:17

more highflying than High Street,

though. As long as we have had PCs,

0:19:170:19:22

we have had flight simulators. But

if you are really serious about

0:19:220:19:26

learning how to fly, then you need

an aircraft and a human pilot to

0:19:260:19:30

teach you what to do. But this lab

is about to be transformed into an

0:19:300:19:35

aircraft cop that, with the help of

this augmented reality headset. And

0:19:350:19:38

when I put it on, it will also

provide me with my very own virtual

0:19:380:19:43

captain. Called Project Cap, it is a

collaboration with aerospace giants

0:19:430:19:52

Boeing.

Necessary information. Sure,

I have doubled my flow. Give it your

0:19:520:19:58

best try.

The aerial cockpit is

designed as a training simulator for

0:19:580:20:04

pilots. They can brush up on skills

or practice in almost any

0:20:040:20:08

environment. It does feel as if I

can reach out and touch the

0:20:080:20:12

controls, and I am very much tempted

to. And I do that, and of course

0:20:120:20:16

there is nothing there. The net. OK,

cap, you seem to have a good idea of

0:20:160:20:24

what to do in this aircraft. So

taxi.

Roger, you are clear to start

0:20:240:20:29

number two.

Ready. At the moment,

Cap response to a very small number

0:20:290:20:35

of voice commands or questions.

Beacon.

On.

Might check.

My cheque,

0:20:350:20:44

one, two on three.

But this can

still be useful for training pilots.

0:20:440:20:49

Augmented reality gives us a chance

to bridge the thing strapped on the

0:20:490:20:53

digital world and the things around

us. How can we start to merge those

0:20:530:20:56

things together in effective ways?

How can we create holograms right

0:20:560:21:00

before you for things that would be

less safe if you were to do them in

0:21:000:21:04

the real world, or that you might

need additional information besides

0:21:040:21:07

what you can build around you in the

real world?

Closed and locked. Looks

0:21:070:21:11

behind him to check that.

VR 140.

It

is a very convincing illusion that

0:21:110:21:18

there is a pilot in here with me.

Any questions?

Do we have a specific

0:21:180:21:25

altitude restriction?

Per year in

development, Cap is actually

0:21:250:21:29

modelled on a real pilot.

We have an

opportunity to take some of our

0:21:290:21:34

friends who are pilots, in this case

one in particular, and see if he

0:21:340:21:37

would actually subject himself to a

full body scan, the then be able to

0:21:370:21:41

use him as our avatar. So that is

who we have, and actual pilots, who

0:21:410:21:45

knows the mannerisms and gestures,

that we can put into that virtual

0:21:450:21:50

pilot's seat.

But is this another

instance of technology putting

0:21:500:21:54

people out of their jobs?

No, not at

all. It is to provide student pilots

0:21:540:22:01

with the opportunity to practise

interpersonal skills before they

0:22:010:22:05

actually get to a flight training

centre, with real pilots. And we can

0:22:050:22:09

provide them with a greater breadth

of experiences, through introducing

0:22:090:22:12

different variables, such as

different cultural types of

0:22:120:22:15

personality styles that they can

practice with.

I do wonder about

0:22:150:22:19

other applications for this sort of

kit. Somebody that might be able to

0:22:190:22:24

teach you how to drive a car, for

instance, will teach you how to

0:22:240:22:28

operate various bits of equipment

and machinery.

Some of the work that

0:22:280:22:32

we had done before doing the work

with Boeing was in things like

0:22:320:22:37

medical simulation, being able to

have a holographic overlays that you

0:22:370:22:40

could see the x-rays laid on top,

exactly placed, or the CT scans or

0:22:400:22:44

MRI is. Those are things that we

think hold great promise, not only

0:22:440:22:48

just because they will help with

visualisation, but they might also

0:22:480:22:51

lead to better quality of care, or

lifesaving, because you have better

0:22:510:22:55

access to data right when you need

it.

So one day, beyond the cockpit,

0:22:550:23:01

Cap's digital descendants might help

teachers teach us how to do all

0:23:010:23:05

kinds of things. -- might help teach

arts. And from Boeing to boosting. I

0:23:050:23:15

am on my way to the Tipsy Robots,

where mythology has been given a

0:23:150:23:23

hi-tech makeover -- boozing. Here,

the drinks are shaken and served by

0:23:230:23:29

these two chaps. I can even invent

my own cocktail by choosing from

0:23:290:23:33

some of the 120 odd spirits hanging

from the ceiling, or I assume all of

0:23:330:23:39

the 120 odd spirits, in one. Can I

do that? No, I can't do that,

0:23:390:23:44

apparently. These droids can mix 100

cocktails are now between the two of

0:23:440:23:49

them. That sounded impressive until

I discovered some human bartenders

0:23:490:23:53

can do ten times that. And that is

it for Click in the US for this

0:23:530:23:58

week. Don't forget, you can follow

us on Twitter, where you can see

0:23:580:24:02

loads of extra backstage videos and

photos, although trust me, you don't

0:24:020:24:05

want to see what happens after I

have one or two of these. Cheers,

0:24:050:24:09

see you soon.

0:24:090:24:11

Download Subtitles

SRT

ASS