– and also a great illustration of how driverless vehicles (at least in a country like the UK) are a distant dream….

Or not. And yes, I know this isn’t the UK but honestly, it’s the countries that seem like they would be impossible for machines to navigate that are the most in need of autonomous driving.

https://www.raillynews.com/2022/05/If-you-get-in-without-a-driver%2C-e-Attack-has-started-to-carry-passengers-in-Norway./

And to be brutally honest, given the standard of the human bus drivers around here, the sooner the current crop are in another line of work the better.

Cougar
Full Member

Trains still need drivers and they’re on rails. Sort that one out first and then get back to me about cars.

Trains need staff, they don’t necessarily need permanent drivers. TBF they’d probably be better with driverless trains and with the driver released from the plate to work the rest of the train.

Of course, triain companies usually see driverless/guardless/inspectorless trains as a way to cut costs and staffing, or to use less qualified staff, but it doesn’t have to be like that. Unions get criticised for fighting it all on those terms, but that’s basically because that’s how it is.

TBF they’d probably be better with driverless trains and with the driver released from the plate to work the rest of the train.

So that’d be the guards and conductors that’ve been the source of numerous strike actions over the last five years, causing massive inconvenience, citing ‘passenger safety’ as a concern, but who hid in the end carriages throughout Covid rather than enforcing basic safety protocols, and almost never IME confront the routine antisocial behaviour that discourages so many from using trains?

That’s like saying we shouldn’t have computers flying planes because my laptop keeps crashing. The tech that’ll go into driverless cars will be completely different to adaptive cruise control

Well, no. The first is comparing planes with laptops, the second is comparing car automation with car automation. If the tech going into driverless cars is going to be “completely different” from current technologies then why can’t we have those technologies instead? Surely the stuff we have now is a precursor to what comes next.

Is your car a VW by the way? My Hyundai has never done anything weird with regards the cruise feature.

Honda. It’s brilliant on motorways but it can’t cope around town. I guess that’s not really what it’s designed for.

IMO the driver of this change are tech companies set to make a fortune. See Tesla.

Free public transport is probably a cheaper and more immediate ‘win’ societally and environmentally.

So many people when they imagine the nirvana of robot cars and no private car ownership don’t live in rural places…

tjagain
Full Member

Once driverless cars become significantly safer than human driven one then it will become harder ad harder to justify driving as you put everyone else at risk

Why? Trains are already far safer than cars but there is no push to ban cars on safety grounds where the train is a realistic alternative.

When you look at absolute risks car deaths are low risk. If we are going to be justifying modes of transport on a risk basis then the first thing to ban is motorcycling.6 miles on a motorbike carries the same risk as 250 miles in a car.

https://plus.maths.org/content/os/issue55/features/risk/index

So many people when they imagine the nirvana of robot cars and no private car ownership don’t live in rural places…

I think autonomous buses will be common long before autonomous cars.

Of course, at some point the line between buses and cars will start to blur. Some of the autonomous buses they have been using here in Norway are probably smaller than a ten seater minivan. The newer ones are much bigger and are used for busier routes in the centre of town.

The beauty of it is that you no longer have to make all your buses 50+ seaters because you won’t have the same issues with training and retaining drivers that you currently have.

When you look at absolute risks car deaths are low risk. If we are going to be justifying modes of transport on a risk basis then the first thing to ban is motorcycling.6 miles on a motorbike carries the same risk as 250 miles in a car.

It’s not clear from your link whether it’s all deaths caused by cars or just the deaths of the occupants.

If you are driving a car then you’re pretty safe. It’s all those pedestrians, cyclists, and motorcyclists who tend to get killed.

irc
Free Member

When you look at absolute risks car deaths are low risk. If we are going to be justifying modes of transport on a risk basis then the first thing to ban is motorcycling.6 miles on a motorbike carries the same risk as 250 miles in a ca

Sure but that’s not a good comparison, because so much motorbike use isn’t transport. You need to separate out recreational use, which is much more common for motorbikes, and also riskier than transport use. You’d see a very different number if you have pure transport use.

Surely the stuff we have now is a precursor to what comes next.

Only in the same way that a Sinclair Spectrum is a precursor to a Macbook Pro. Your (and my) adaptive cruise is noddy, only has one sensor of one type, and is not certified to do anything at all. A driverless car will be covered in loads of visual cameras and lidars, have far more computing power, and 1000x the lines of code, and it will have to go through a rigorous certification process.

So many people when they imagine the nirvana of robot cars and no private car ownership

So many people who live in rural places don’t realise that they are seriously in the minority and think that their exceptions invalidate the whole idea.

But besides that, robot cars are a different thing to no car ownership, they are not interdependent. Robot cars just make better taxi drivers, that’s the only overlap.

Only in the same way that a Sinclair Spectrum is a precursor to a Macbook Pro. Your (and my) adaptive cruise is noddy, only has one sensor of one type, and is not certified to do anything at all. A driverless car will be covered in loads of visual cameras and lidars, have far more computing power, and 1000x the lines of code, and it will have to go through a rigorous certification process.

yep. Moore’s law innit. The fields of AI/Machine Learning are advancing at a stupendous rate currently, and that rate will only accelerate. Stuff that would’ve been science-fiction only a few years ago, or at the very least very difficult/expensive to implement, is now commonplace. This is the only thing holding back driverless cars now… everything else, physical control methods, etc, was sorted ages ago.

zilog – the most difficult thing I thought was the decision making – ie kill the granny on the pavement or the cyclist in the road – that sort of issue along with making decisions in a rapidly changing environment

I think its a great idea…one big issue I think, is that insurance for human piloted cars might become very expensive as they would be a much higher risk.
Although I suppose people really out in the sticks might be able to buy a specialist policy with limited millage and only for driving to an ‘auto car hub’?

So many people who live in rural places don’t realise that they are seriously in the minority and think that their exceptions invalidate the whole idea.

You could apply that more widely though, e.g people who want the advantages of rural life without disadvantages like poor broadband and inadequate public transport. Suck it up. Regarding the original question as to who wants driverless vehicles – the corporations who jumped on a concept that seemed like a money tree a decade ago but are now realising it might be a financial black hole they’ve been pouring their money into?

the most difficult thing I thought was the decision making – ie kill the granny on the pavement or the cyclist in the road

actually implementing the algorithm is trivial… it’s more of a moral conundrum which way the computer should be taught to go! I’ve seen this posed as a “thought exercise” re. autonomous cars, but I think maybe people make too big deal out of this… often the scenarios posed just wouldn’t be faced by the computer-controlled vehicle, because they wouldn’t have made the same mistakes as a human to get them into that pickle in the first place!

So many people when they imagine the nirvana of robot cars and no private car ownership don’t live in rural places…

So many people who live in rural places don’t realise that they are seriously in the minority and think that their exceptions invalidate the whole idea.

I live in a rural area which has two busses per week (Mon 10-14:00 and Thursday 11:15 to 14:45) each to A local town 8.5 miles away. You have to do whatever you want to do and be back on the bus in 4 hours in the middle of the day, in the middle of the week.

A self driving, self recharging electric car service for the village with 1 or 2 or more available cars with 4-6 seats would be FAAR more useful for everyone in our village and those around us.

I think its a great idea…one big issue I think, is that insurance for human piloted cars might become very expensive as they would be a much higher risk.

They would be no higher risk than now. Possibly lower. If EVs, as claimed, are safer then there would be less 50:50 bumps. So insurance for human drivers would be lower with less accidents overall.

What it might do is reduce the number of new drivers. A new young driver currently pays huge insurance until older and more experienced. The prize at the moment is the mobility provided by the car.

If there is a way to get virtually the same mobility without paying several years of young driver insurance then a licence is a bit less attrative.

Meanwhile my annual insurance, under £200 is a small part of overall costs and won’t influence my car choices.

Sure but that’s not a good comparison, because so much motorbike use isn’t transport. You need to separate out recreational use, which is much more common for motorbikes, and also riskier than transport use. You’d see a very different number if you have pure transport use.

Different but still very risky. Motorbikes are under 1% of traffic but 20% of road deaths.
66% of motorway deaths are rural. Call all rural miles recreational and it leaves a third of urban miles for transport/utitlity. Motorbiking is still around 7 times riskier than average.

UK Motorcycle Accident Statistics: The Complete 2022 Guide

So as I said if we want to go Big Brother and start banning modes of transport on a risk basis bikes will be first.

set it to “Alps”

Make sure you don’t arrive on a snowy day. Just one of many examples when you’ll end up walking to your destination when the conditions mean the car decides to stop or does something stupid where a driver would make do with limited visibility and grip.

Sounds like a strong argument for driverless cars.

It’s going to be lovely. I used to love driving in rural England, but here in Vancouver it’s a real chore. Traffic is slow, people are inconsiderate and impatient, single occupancy vehicles (usually gigantic pickup trucks) clog the roads. I can’t wait until I don’t have to worry about Chad trying to merge onto my bonnet in his F-350 to get over to the McDonalds drive-through.

In my mind it’ll happen slowly, maybe with people who can afford the initial sum buying full self drive cars, but then leasing them out when they’re not using them….I suppose like an Air B&B.

There’s a car share program here that I use a lot, called Evo, and it’s great. All the cars are electric or hybrid, generally they’re clean and tidy, don’t smell, they all have 2 bike racks and a ski rack, and you just file a short report when you get in (is it clean? Any new damage? Etc). There’s tons of cars, you can park them pretty much anywhere for free, and you pay buy time. It’s about 25p per min or £10 an hour. Just park it up and leave it for the next person. Next logical step seems to make them self driving, I suppose. You could even select if you wanted to “ride share” and pick people up on the way.

I’d rather we just got on with making cars obsolete TBH.

Make sure you don’t arrive on a snowy day. Just one of many examples when you’ll end up walking to your destination when the conditions mean the car decides to stop or does something stupid where a driver would make do with limited visibility and grip.

Sounds like the cars the sensible one there.

How many cyclists have died because drivers decided they “would make do with limited visibility and grip.”?

So many people when they imagine the nirvana of robot cars and no private car ownership don’t live in rural places…

This always gets trotted out, usually (although not in Matt’s case, IIRC he works in outdoor education?) by middle aged, middle class people who’ve chosen to live there and accept their 1h commute into “the big smoke”.

Without a hint of irony that their situation is self inflicted and self fulfilling because they keep driving into the cities.

Ban the cars from the cities however and most of the problem goes away because their commute becomes too difficult. With the knock on effect of cities then become nicer places to live so peole are less inclined to leave in the first place. And rural house prices cease to spiral out of controll with fewer commuters competing with people who actually “live” and work there.

^ City vs rural won’t be so much an issue- it will be these growing sprawling suburban/semi-rural developments (not new towns, just ‘developments’) Best guess is with little to no multimodal infrastructure because cars are the normal/future of UK transport and automation/wfh/gig economy will be the lifestyle. Cities will
be the few places where there will be some (minimal/poorly-implemented) cycling infrastructure.

If the tech going into driverless cars is going to be “completely different” from current technologies then why can’t we have those technologies instead? Surely the stuff we have now is a precursor to what comes next.

Because cars need to be able to image and sensor process, compute velocities, vectors, stopping distances, conditions, reserve power, steering angle, road inclination, vehicle dynamics, etc and from all that, create a hierarchy of possible decisions based on allowed laws and safety considerations in a dynamic environment. It must do all that without ANY internet connectivity, all on-board, within a split second on a low power consumption, low cost, fully connected, totally stable computer system. THIS is the challenge to fully autonomous vehicles.

I think we need to be sensible, if your are driving up mountains etc, a regular car might be a better option, and exeptions can and should be made.

But there are litteraly billions of journeys per day to tesco, or for the school run, or the work commute that can instantly be made far safer and far better for the environment, by removing the human driver aspect.

let’s tackle the problem from where it is worst. We’re not taking about stopping farmer Giles in his Suzuki Jimney from using his utility vehicle.

@p7eaven, yup. Arround Reading it’s just nuts.

There’s thousands of new flats going up in the town center for people who will commute by train.

Then thousands more of suburban sprawl in “Wokigham” because government targets say so. Why? There aren’t any jobs for these people, they’re just going to be forced to drive to somewhere else to work (or into Reading to park at the multistoreys) . It’s a completely flawed way to build houses!

And because it’s all low density suburban sprawl it’s just making it worse. Housing estates miles across don’t encourage anyone to walk to the bloody shops do they?

So many people who live in rural places don’t realise that they are seriously in the minority and think that their exceptions invalidate the whole idea.

@molgrips – I think my point wasn’t about somehow rejecting ideas, more that rural needs a lot of thought. Both residents and visitors.
I agree with the thought that rural buses will be smaller and possibly more regular.
.
I’ve had the ‘hardship’ today of driving from Aviemore to the Moray coast. It’s the kind of place that’s got one or three houses a mile from a village of 50 houses, which in turn is 8 miles from the town. In winter its icy and snowy. The roads flood. The villagers work in town, at home, the next valley over village school (etc). Currently they can head out for the evening to town, or perhaps majority of the “village” (a whole valley…) gather at once at the village hall.
I’m working here to two days – I’ve got a car as it’s only way of getting between the towns and villages I’m working in.
There’s a line of cars outside the pub tonight, first one with a London garage number plates and a couple of touring bikes on the back.
These things all need a solution, and I’m not yet convinced that technology has it covered, or that rural life is facing a massive change coming.
.
A just transition is going to be difficult.

Weren’t we supposed to have hover cars by now?

you’ll end up walking to your destination when the conditions mean the car decides to stop or does something stupid

I don’t think ‘do something stupid’ will be programmed into the car. It’s programmed into a lot of humans though.

How many cyclists have died because drivers decided they “would make do with limited visibility and grip.”?

In those snowy conditions I cited you’d need a fat bike and goggles with wipers so I suggezst zero. I’d rather trust a driver’s eyes than a load of sensors trying to deal with falling sticky snow.

People trusting driverless cars have already killed pedestrians/cyclists in unsnowy and grippy conditions as you’ll know if you’ve followed the previous threads on here.

Alert driver versus driverless car, I’ll take my chances with the driver.
Distracted/Angry anti-cyclist driver, I’d rather take my chances with a driverless car.

rural needs a lot of thought

It does. Let’s think about what driverless vehicles would add. I mean, we already have busses and taxis, right?

A taxi driver needs to make say £30kpa to make it worth it, plus they has to buy their car. That means they have to charge fairly significant chunks of money, especially in the countryside where things are further apart. This makes them a less viable option since demand is suppressed by this price/cost floor. That then suppresses the supply which means it takes longer to order a taxi and since taxi drivers have to go home at night they have a.limit on how far they’ll go. All a bit of a faff and expense.

However a driverless car with enough range could cost in today’s money £30k, and let’s assume all the driverless gear adds ten grand. It could easily run for a decade on that cost, like the human driver’s car, but you’re saving £300k by not paying the driver. That could make it easy to keep the fare down. It might be as cheap as bus fare, but it goes door to door, is quicker and private. That could increase demand (I’d have used this constantly as a youngster, as it was I had to hitch hike or ride) and you could just keep putting more and more on the roads and they could become hugely important for rural communities. But wait, what about congestion? Well, there’s not much of that out in the real countryside now, and the driverless cars wouldn’t necessarily have to drive home again empty after a dropoff so overall car miles cod actually be less.

Add a bit more tech and you could car pool easily. Tell it you need to go to town some time this week and it’ll offer you a trip rolled up with other people and you’d save even more. Like an on-demand bus.

People trusting driverless cars have already killed pedestrians/cyclists in unsnowy and grippy conditions as you’ll know if you’ve followed the previous threads on here.

And human drivers kill people all the time.

If I were designing a system I’d have it listen to weather and conditions reports from the resort and other cars on the road, and if they were bad you’d get a message saying ‘hey, conditions on the road are terrible, I’m not going up there in this. You’re on your own.’ Then you drive up anyway on manual controls being a driving god of course. Good luck trying to make an insurance claim if anything were to happen though.

Can’t wait to see a driverless car/vehicle try & get across the Empress roundabout in Harrogate at 4.30pm on a Friday. It would be carnage*.
Not unless every other vehicle there is also driverless.

*bussage? vanage?

I have confidence in the tech but you will be able to bully driverless cars out of the way by relying on them stopping, I suspect. I don’t think there’s a way around that. Unless it simply uploads footage of you being a dick to the police AI that automatically sends you a ticket.. hehe…

Add a bit more tech and you could car pool easily

Let’s be honest, if we weren’t all so precious about our cars, we could car pool and lift share already!

I agree with the thoughts about sprawling towns without ‘local level’ infrastructure such as high street shopping in walking distance, local schools and buses or trains to other destinations.

I still think the model of a driverless car on a Highland road is a hell of a long way away – and the driverless £100k Tesla to take me on my old commute rather than my battered old car doesn’t initially stack up financially.

I’m also aware I don’t want to conflate driverless tech with low carbon tech or social solutions. All three seem to be an overlapping Venn diagram – but also different issues…

I still think the model of a driverless car on a Highland road is a hell of a long way away

probably overthinking the importance of Highland roads in the UK transport network as a whole tbh 😂

I suspect they’ll start with the centres of the big cities, kind of like ULEZ, where either you can only drive an autonomous car or they’ll be a charge to drive a non-autonomous one, and move outwards from there. By the time they get to you you’ll probably be able to pick up an ancient autonomous banger for a weeks wages 😃

Then you drive up anyway on manual controls

That assumes you have the choice, give people the choice and I think they’ll make quite selective use of self-driving options and all the dangers of human operated cars will reamin with some new ones due to driverless technology.

Tesla might have some stats about whether drivers or the auto pilot screws up more in conditions in which the self-driving mode is appropriate but I can’t see them releasing uncensored data to anyone.

I very much doubt we will see properly autonomous private cars in my lifetime. I think we are decades away yet. the experiments in the states show how far we are away and there is still a huge issue around liability. Who is liable if an autonomous car kills someone?

We might see limited use before that like the buses mentioned above – that run a fixed urban route and perhaps some motorway driving. But a car you can sit in in london and tell it to take you to Aviemore? decades away IMO

Also they really do not solve the main issue of energy usage. I suspect the private car will be priced out of the reach of most of the population first

To give another answer to the original question, tech companies like Google are pushing it because people in automated vehicles are a captive market with nothing better to do than scroll their phones and consume more stuff. Imagine how much more you would use stw if you didn’t have to look at the road on the way to work