73% of Brits would not feel safe with driverless cars on the road

James2k

unholyangel wrote: »

Sorry, my wording was abysmal. I meant they already step out into the road when they perceive less risk. You're more likely to get people stepping out at crossings, traffic lights, in slow moving traffic etc because they perceive less risk. So in other words, the less risk there is, the higher number of people you'll get willing to do it.

yeah this is true

welfayre

Can't say I'm bothered either way about them but I'm really looking forward to the government trying to explain how we can't have driverless trains and buses for safety reasons but driverless cars are fine.

Ergates

Richard53 wrote: »

Until the moment that they realise that car makers are terrified of being the cause of an accident (imagine the publicity) and that the car will do virtually anything to avoid running them over. After that, it's game on.


Now translate that the normal road network, like round your house or my village. Are the pedestrians, cyclists and dogs going to get chips implanted so they will interact 'correctly' with the motor traffic? Only in some sci-fi nightmare.

What's to stop people doing that right now? Most drivers would also do pretty much anything to avoid running over a pedestrian.

I get the sense you're projecting a little.

zagfles

unholyangel wrote: »

The luddites thought that having everything (as opposed to in moderation) automated could be problematic for the economy? Cool story bro.

They didn't have the imagination to envisage the new opportunities of increasing automation, they only saw it as taking their livelihoods away.

But at least they had the excuse of not having 200 years of history to look back on. Unlike their equivalents in recent times who've been predicting exactly the same with every new technology.

Sea_Shell

You only have to look at the Aviation industry and watch a few episodes of "Air Crash Investigation" to know that Autopilot, whilst a brilliant aide, is not infallible, and a human pilot sometimes has to intervene to save the plane from itself. In a lot of cases, the planes crash because the pilot DIDN'T overrule the automation early enough, as it was perceived that it can't be wrong!!!

marlot

One thing I'm looking forward to is many fewer parked cars.


Lots of our city streets were designed before cars became so universal.


If you can just summons a car when you want to, then people will (hopefully) move away from owning cars and the streets will become much clearer again.

marlot

welfayre wrote: »

Can't say I'm bothered either way about them but I'm really looking forward to the government trying to explain how we can't have driverless trains and buses for safety reasons but driverless cars are fine.

I've been on driverless trains in the UK. As have about 340,000 people a day. https://en.wikipedia.org/wiki/Docklands_Light_Railway

Trains are the easiest transport to automate because the tracks take care of the navigation and its accepted that pedestrians have to keep off the track.

marlot

It's pretty shocking that as a society we put up with almost 2,000 people dying on our roads each year, and about 25,000 being seriously injured.


People are allowed to pass a driving test at 17, and never have to have a moment of top-up training in their life.


I was driving on the M4 yesterday, and there was a whole stream of traffic at 70+ driving less than 30 feet from the car behind. Along with a mixture of others clearly on their phones.


I'm not looking for driverless cars to make the roads 100% safe (though that would be great). Just getting them safer than today would be acceptable.


I do wonder whether the most dangerous time is the intorductory phase - when we have early driverless cars (with incomplete technology) mixing with driver-operated cars. At some point, I'd expect driver-operated cars to be banned or restricted to one lane of the motorways.

AdrianC

marlot wrote: »

It's pretty shocking that as a society we put up with almost 2,000 people dying on our roads each year, and about 25,000 being seriously injured.

Just to put that into proportion, "seriously injured" is officially defined as...

An injury for which a person is detained in hospital as an “in-patient”, or any of the following injuries whether or not they are detained in hospital: fractures, concussion, internal injuries, crushings, burns (excluding friction burns), severe cuts,severe general shock requiring medical treatment and injuries causing death 30 or more days after the accident. An injured casualty is recorded as seriously or slightly injured by the police on the basis of information available within a short time of the accident. This generally will not reflect the results of a medical examination, but may be influenced according to whether the casualty is hospitalised or not. Hospitalisation procedures will vary regionally

So when SWMBO was walking through a park a couple of years ago, and was hit by a cyclist, resulting in a cut scalp and concussion - but just a visit to A&E - that was "seriously injured".


The most recent official statistics - https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/744077/reported-road-casualties-annual-report-2017.pdf
1,793 fatalities - 39% reduction on 2007, but no real change from the 2010-2014 average.
24,831 serious injuries - not comparable to previous years, due to reporting changes.
So, yes, the figures are about right.

BUT...
Look at the rate per distance travelled.
238 casualties per BILLION person-miles by car, 1.9 deaths.
1,801 casualties per BILLION person-miles by foot, 35.6 deaths.
The single most hazardous group? Motorcyclists.
6,043 casualties per BILLION person-miles, 116.9 deaths.



And that's BILLION person-miles... How far do you walk in a year? How far do you travel by car?
If you walk a mile a day, your chances of being killed are one in 77,000.
If you travel 15,000 miles/year by car, your chances of being killed are one in 35,000

That's pretty damn low...

And even that ignores what percentage of those are the author of their own misfortune. We had a near-fatal collision just up the road last year. Far too fast, ran out of road, and a fence post through the windscreen. I can think of two fatal collisions near here in recent years. One was drunk, far too fast and went straight on at a bend into a tree. The other was elderly, and pulled straight out of his drive in front of an articulated wagon.

So let's be generous and say half are own-stupid-fault.
So that's now a 1 in 70,000 chance of being killed in a car, and 1 in 154,000 on foot.


Your annual risk of needing emergency treatment after being injured by a mattress or pillow is one in 2,000...

People are allowed to pass a driving test at 17, and never have to have a moment of top-up training in their life.

Yep, regular re-testing would be a very good idea.

I was driving on the M4 yesterday, and there was a whole stream of traffic at 70+ driving less than 30 feet from the car behind. Along with a mixture of others clearly on their phones.

There should be a law against it!

I'm not looking for driverless cars to make the roads 100% safe (though that would be great). Just getting them safer than today would be acceptable.

Define "safer than today".
Now think about the liability issues for the autonomous car manufacturers when there is an incident...

I do wonder whether the most dangerous time is the intorductory phase - when we have early driverless cars (with incomplete technology) mixing with driver-operated cars. At some point, I'd expect driver-operated cars to be banned or restricted to one lane of the motorways.

The big problems with autonomous cars are currently the "out-of-the-ordinary" situations. How does an autonomous car deal with taking you to the site of a big temporary event? There might be informal carparks springing open in fields. Official carparks might be full or closed because of adverse weather. There may be a park and ride a mile or two away. How does the car know any of this? The main road through our village has been closed for several days this week - initially because of emergency roadworks for a water leak, then for removal of a fallen tree that took down phone cabling. How does it deal with that? The signed diversions took people down single-track lanes. How do they deal with having to reverse to a passing place? Or meeting somebody on a horse down that lane?



If and when somebody in an autonomous car arrives at my house, how does it know where in the drive to park so that it won't block the public footpath or garage? Or not to park under that tree, because there's still some hung-up bits of branch from a recent trim?

And that all assumes the car can talk to "base"... We have zero mobile phone signal round here - not even send-and-receive SMS, let alone any viable data connection.

zagfles

AdrianC wrote: »

Just to put that into proportion, "seriously injured" is officially defined as...


So when SWMBO was walking through a park a couple of years ago, and was hit by a cyclist, resulting in a cut scalp and concussion - but just a visit to A&E - that was "seriously injured".


The most recent official statistics - https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/744077/reported-road-casualties-annual-report-2017.pdf
1,793 fatalities - 39% reduction on 2007, but no real change from the 2010-2014 average.
24,831 serious injuries - not comparable to previous years, due to reporting changes.
So, yes, the figures are about right.

BUT...
Look at the rate per distance travelled.
238 casualties per BILLION person-miles by car, 1.9 deaths.
1,801 casualties per BILLION person-miles by foot, 35.6 deaths.
The single most hazardous group? Motorcyclists.
6,043 casualties per BILLION person-miles, 116.9 deaths.



And that's BILLION person-miles... How far do you walk in a year? How far do you travel by car?
If you walk a mile a day, your chances of being killed are one in 77,000.
If you travel 15,000 miles/year by car, your chances of being killed are one in 35,000

That's pretty damn low...

Yet when a few tens are killed in a terrorist attack, or Grenfell, or a stadium disaster etc, there is a national meltdown. Yet thousands getting killed on the roads, or tens of thousands by pollution, well that's just OK because the risk is "low" of it happening to you.

And even that ignores what percentage of those are the author of their own misfortune. We had a near-fatal collision just up the road last year. Far too fast, ran out of road, and a fence post through the windscreen. I can think of two fatal collisions near here in recent years. One was drunk, far too fast and went straight on at a bend into a tree. The other was elderly, and pulled straight out of his drive in front of an articulated wagon.

So let's be generous and say half are own-stupid-fault.
So that's now a 1 in 70,000 chance of being killed in a car, and 1 in 154,000 on foot.


Your annual risk of needing emergency treatment after being injured by a mattress or pillow is one in 2,000...



Yep, regular re-testing would be a very good idea.



There should be a law against it!



Define "safer than today".
Now think about the liability issues for the autonomous car manufacturers when there is an incident...



The big problems with autonomous cars are currently the "out-of-the-ordinary" situations. How does an autonomous car deal with taking you to the site of a big temporary event? There might be informal carparks springing open in fields. Official carparks might be full or closed because of adverse weather. There may be a park and ride a mile or two away. How does the car know any of this? The main road through our village has been closed for several days this week - initially because of emergency roadworks for a water leak, then for removal of a fallen tree that took down phone cabling. How does it deal with that? The signed diversions took people down single-track lanes. How do they deal with having to reverse to a passing place? Or meeting somebody on a horse down that lane?



If and when somebody in an autonomous car arrives at my house, how does it know where in the drive to park so that it won't block the public footpath or garage? Or not to park under that tree, because there's still some hung-up bits of branch from a recent trim?

And that all assumes the car can talk to "base"... We have zero mobile phone signal round here - not even send-and-receive SMS, let alone any viable data connection.

Err, the passenger "tells" it? Just like the passenger would tell a driver where to go, where to park etc if they know the area better. The mechanism of "telling" may be different, but same principle.