How should we make the transition from driving to driverless vehicles in the future?

[AM9]

I wonder if trucks in a platoon will show some kind of light to indicate that situation. In other words, how would a car driver know that they have joined in the middle of a platoon ? Would there be any legal obligation on a car driver to change lane and allow the platoon to re-form ?

All these complicationns so that truck drivers can text their friends on the move.

 

[biko]

It doesn't explain how its done. Does the car have to shove its way into the subsafe gap and be unsafe until the truck has slowly adjusted?

I think this is still very much under development. Papers that I saw about this suggest the trucks should automatically register oncoming cars and decide between which trucks a gap should be left.

Both of these things will lead to platoons overtaking or being overtaken by other trucks -the thing that really really increases congestion

I agree that should not be the case, but I believe that isn’t the main goal of these tests. In the first step this is required as practically no vehicles are equipped with this technology and mostly to test the technology.

Also allowing cars into the middle of the platoon might have legal impacts on whether the following drivers are driving or not

Yes, that’s complex and part of research as I understand it.

 

[Jozhua]

All these complicationns so that truck drivers can text their friends on the move.

Many already do

 

[philthetube]

I do wonder how driverless cars would cope with the latest alterations to the highway code concerning pedestrians?

 

[Bletchleyite]

I do wonder how driverless cars would cope with the latest alterations to the highway code concerning pedestrians?

Far better than manually driven ones which appear thus far to be ignoring it entirely.

 

[Meerkat]

I do wonder how driverless cars would cope with the latest alterations to the highway code concerning pedestrians?

Terribly, as pedestrians will know they can stop the driverles cars and without a human edging the car forward there will be no human interaction and compromise.
On the other side of the coin slow pedestrians (eg the elderly or parents with pushchairs) will wait ages to cross roads where human drivers would stop and wave them across.
Many urban road systems would need expensive changes if driverless cars are introduced without central control - I can think of numerous junctions which would create gridlock if cars didnt push in and/or be let in by the cars with priority.

 

[Bletchleyite]

Terribly, as pedestrians will know they can stop the driverles cars and without a human edging the car forward there will be no human interaction and compromise.

There isn't supposed to be compromise in this new rule. Effectively, a junction is a zebra crossing in its effect with absolute pedestrian priority.

 

[Meerkat]

There isn't supposed to be compromise in this new rule. Effectively, a junction is a zebra crossing in its effect with absolute pedestrian priority.

Sure, but things work because generally humans are reasonable people. Some will march out but others will pause a second and let a waiting car through.
Some places this will have to happen or cars waiting to turn will block main roads for ages.

 

[AM9]

Sure, but things work because generally humans are reasonable people. Some will march out but others will pause a second and let a waiting car through.
Some places this will have to happen or cars waiting to turn will block main roads for ages.

But some humans aren't generally reasonable people, and (perish the thought) when in their cars they rarely respect the safety and convenience of others from the safety of their vehicles. That's been observed for years with the sad records of many incidents on the road where pedestrians have paid a high or even the ultimate price for that lack of resonableness. The change in the law is a serious attempt to educate drivers and hopefully save some lives. Clearly the views of some are going to make that a difficult and painful task.

 

[Bletchleyite]

Some places this will have to happen or cars waiting to turn will block main roads for ages.

That's the exact point the change is to make - other than on motorways and special roads where they are banned, pedestrians are more important than cars (both for environmental and safety reasons), and thus cars give way to pedestrians even if it happens to cause them or other cars delay.

In urban areas that seems right to me. It's been the way Germanic countries have tended to do it for a long time.

 

[Meerkat]

But some humans aren't generally reasonable people, and (perish the thought) when in their cars they rarely respect the safety and convenience of others from the safety of their vehicles. That's been observed for years with the sad records of many incidents on the road where pedestrians have paid a high or even the ultimate price for that lack of resonableness. The change in the law is a serious attempt to educate drivers and hopefully save some lives. Clearly the views of some are going to make that a difficult and painful task.

That's the exact point the change is to make - other than on motorways and special roads where they are banned, pedestrians are more important than cars (both for environmental and safety reasons), and thus cars give way to pedestrians even if it happens to cause them or other cars delay.

In urban areas that seems right to me. It's been the way Germanic countries have tended to do it for a long time.

With respect I think you are both forgetting the title of this thread. I am not disagreeing with the changes to the Highway Code, I am just saying they make it even harder for a transition to autonomous vehicles (unless there is central control which is even more unlikely).
I commuted by car through suburbia and after pondering autonomous vehicles I really began to notice how often the reasonableness and good nature of drivers and pedestrians kept the traffic flowing and the pedestrians moving - people with the right of way waiving that right to help others. Without it many areas would be gridlocked, and pedestrians stuck for ages or walking much further to official crossings.

 

[AM9]

With respect I think you are both forgetting the title of this thread. I am not disagreeing with the changes to the Highway Code, I am just saying they make it even harder for a transition to autonomous vehicles (unless there is central control which is even more unlikely).
I commuted by car through suburbia and after pondering autonomous vehicles I really began to notice how often the reasonableness and good nature of drivers and pedestrians kept the traffic flowing and the pedestrians moving - people with the right of way waiving that right to help others. Without it many areas would be gridlocked, and pedestrians stuck for ages or walking much further to official crossings.

Experience to date of some motorists' behaviour at traffic lights, pelican crossings and zebra crossings shows that most motorists stop because it is the law, but when some don't the outcome is usually worse for the pedestrians. These changes in priorities are the law now, so they should be regarded the same as zebras etc. without drivers or autonomous vehicles' programs having a 'reasonable' option.
If there were road features that relied on pedestrians being 'reasonable' by waiving their right of way which have been determined for their safety anyway, surely the road layout and signage/traffic signals would be not fit for purpose and due for modification. To perpetuate a situation for pedestrians where the hazards it presents can only be mitigated by them having to waive their right of way, makes a mockery of claims to put the most vulnerable road users' safety at the highest priority.

 

[Meerkat]

Experience to date of some motorists' behaviour at traffic lights, pelican crossings and zebra crossings shows that most motorists stop because it is the law, but when some don't the outcome is usually worse for the pedestrians. These changes in priorities are the law now, so they should be regarded the same as zebras etc. without drivers or autonomous vehicles' programs having a 'reasonable' option.
If there were road features that relied on pedestrians being 'reasonable' by waiving their right of way which have been determined for their safety anyway, surely the road layout and signage/traffic signals would be not fit for purpose and due for modification. To perpetuate a situation for pedestrians where the hazards it presents can only be mitigated by them having to waive their right of way, makes a mockery of claims to put the most vulnerable road users' safety at the highest priority.

Again someone is missing the point!!
I am not arguing whether the changes were necessary or not, just their impact on trying to keep the traffic and pedestrians moving if cars were autonomous.
You also missed that its a two way street and not an attack on pedestrians - plenty of cases where drivers stop (when they dont have to by law) so that pedestrians can cross.
I can't see them programming an autonomous car to think "that old dear is going to be waiting ages for a gap in traffic to cross the road, I'm going to stop and let her cross"

 

[Ediswan]

These changes in priorities are the law now

I am aware that there are new Highway Code rules, but I have not noticed any changes to the law.

 

[Bletchleyite]

I am aware that there are new Highway Code rules, but I have not noticed any changes to the law.

They appear to be "should"s and thus have no legal weight, other than for charges like driving without due care and attention and (causing death by) dangerous driving, where failure to follow them would be contributory.

 

[Meerkat]

They appear to be "should"s and thus have no legal weight, other than for charges like driving without due care and attention and (causing death by) dangerous driving, where failure to follow them would be contributory.

Also affects civil cases so if you survive getting run over you should get more compo!

 

[AM9]

Again someone is missing the point!!
I am not arguing whether the changes were necessary or not, just their impact on trying to keep the traffic and pedestrians moving if cars were autonomous.
You also missed that its a two way street and not an attack on pedestrians - plenty of cases where drivers stop (when they dont have to by law) so that pedestrians can cross.
I can't see them programming an autonomous car to think "that old dear is going to be waiting ages for a gap in traffic to cross the road, I'm going to stop and let her cross"

With respect I think you are missing the point. I presume what you mean by 'two way street' is that drivers stop where they don't have to so pedestrians should do the same. We iif you want to think of it like that, bear in mind that it is a very unequal two way street, i.e. if motorists choose not to give way where the HC says that they should, it could cause an injury or worse. The motorist is unlikely to suffer any injury. If a pedestrian chooses to cross where the HC says that motorists should give way and the motorist heeds the recommendation, at worst the motorist is delayed by a few seconds. That's not a two way street in my book. Now of course the motorist's defence would claim in court that the HC doesn't say they must stop, ergo it's all down to the pedestrian being unreasonable. Not a very good look.
Mmmm, if that comes up too often then that 'should' would be changed to a 'must' pretty soon. Beware of what you wish for. The priority change must be seen as a genuine attempt to protect the vulnerable, not an inconvenience for motorists.

 

[philthetube]

That's the exact point the change is to make - other than on motorways and special roads where they are banned, pedestrians are more important than cars (both for environmental and safety reasons), and thus cars give way to pedestrians even if it happens to cause them or other cars delay.

In urban areas that seems right to me. It's been the way Germanic countries have tended to do it for a long time.

What happens on a dual carriageway, people are allowed to cross, would a car in the outside lane see a pedestrian crossing in front of a car on the inside lane, frightening.

 

[PTR 444]

I’ve been wondering, could driverless cars begin life as part of a PRT-like system but using public roads rather than a segregated guideway network?

The Morgantown system is intriguing in that it allows point to point, turn-up-and-go travel without the need for a driver, but is only achievable because it uses a segregated guideway. Now that autonomous vehicles are more likely to become reality, I suggested that a segregated PRT-like system could work for AVs using a normal road surface rather than guideway, but as others have pointed out it would have the disadvantage of not being able to provide door-to-door travel for everywhere, which is one of the main objectives of AVs. Having thought about this further though, could a non-segregated version of a PRT work using autonomous road vehicles in areas of low traffic?

I say low traffic because as mentioned before, introducing AVs to areas of high traffic density will likely cause problems until technology improves vastly to the point that AVs are 100% fail-safe and foolproof. I think they would be better off as a replacement for rural demand-responsive transport, with separate networks using accessible pod-like vehicles to provide travel between a cluster of rural villages and their nearest market town or railway station. These vehicles would travel along the quietest roads wherever possible so as to reduce conflicts with other traffic.

The removal of drivers from the equation brings two advantages over existing DRT. Firstly, costs are reduced so more vehicles can be provided which in turn allows each user/group to have a vehicle to themselves, and secondly bookings can be made for any journey at any time, rather than having a booking deadline on the day before one is due to travel. Both of these would bring DRT more in line with a conventional taxi service, albeit without a driver present so you could argue that it is actually a PRT using public roads rather than a guideway system.

While this would be a godsend for rural areas without any alternative form of public transport, I don’t think it would be workable in urban areas or those which already have a commercially viable bus service. I say this because if you started to introduce driverless DRT to those areas, passengers would be abstracted from existing transport in the area, and the personal nature of this concept means you would have to provide more vehicles, bumping up costs and increasing congestion in the process. It would then become a disaster if manually driven vehicles still existed in those areas, hence why I suggest that fully driverless cars should stick to rural areas and controlled-access roads during the early stages of life.

 

[Bletchleyite]

Small-vehicle public transport is little better than taxis, which are little better than private cars. So no.

 

[PTR 444]

Small-vehicle public transport is little better than taxis, which are little better than private cars. So no.

You could tweak the system so that the vehicles are larger and more than one group can travel in one if they are going to the same destination (even the Morgantown PRT does this), but the idea is that nobody ends up inconvenienced because their journey took longer as a result of the vehicle going off-route to pick up someone else.

 

[Bletchleyite]

You could tweak the system so that the vehicles are larger and more than one group can travel in one if they are going to the same destination (even the Morgantown PRT does this), but the idea is that nobody ends up inconvenienced because their journey took longer as a result of the vehicle going off-route to pick up someone else.

I know, but the problem with private vehicles remains regardless of how they work as private vehicles, i.e. poor energy use and poor use of infrastructure. Self-driving reduces safety risk, but so do driver-assist collision-avoidance systems (which are simpler to do than full self-driving).

There is little reason to favour "public transport" that is just a variation on the taxi except in niche situations like the Heathrow pods. We don't have to eradicate cars or taxis, sometimes they serve a useful purpose.

 

[PTR 444]

I know, but the problem with private vehicles remains regardless of how they work as private vehicles, i.e. poor energy use and poor use of infrastructure. Self-driving reduces safety risk, but so do driver-assist collision-avoidance systems (which are simpler to do than full self-driving).

There is little reason to favour "public transport" that is just a variation on the taxi except in niche situations like the Heathrow pods. We don't have to eradicate cars or taxis, sometimes they serve a useful purpose.

But the idea is that these driverless pods would provide a lifeline to those in rural areas who cannot drive, such as young teenagers and the elderly, in areas where providing a bus service would not be commercially viable. They would be affordable too unlike a taxi which is expensive because it has to cover the cost of the driver.

Unless of course you can automate the bus so enough costs are saved there to make it viable to run a fixed-route, fixed-timetable rural service.

 

[Cloud Strife]

Once ownership is at a high level, the UK motorway network is adapted to allow autonomous driving only as this will increase capacity on such roads.

This would be an absolute pleasure for me on long journeys. I'm dreaming of a situation where I can get in the car, drive to the local motorway, then sit back with a book for a few hours while the car heads towards the Croatian coast. With the exception of one small section between the Czech/Austrian border and south of Brno, there's roughly 1250km to Split, which I do in roughly 12-13 hours. If the car was completely autonomous, I could leave at midnight and arrive by the coast for 2-3pm, which would be perfect.

I like driving overnight, but I can't deny that it's very very difficult once the sun comes up.

 

[Bletchleyite]

This would be an absolute pleasure for me on long journeys. I'm dreaming of a situation where I can get in the car, drive to the local motorway, then sit back with a book for a few hours while the car heads towards the Croatian coast. With the exception of one small section between the Czech/Austrian border and south of Brno, there's roughly 1250km to Split, which I do in roughly 12-13 hours. If the car was completely autonomous, I could leave at midnight and arrive by the coast for 2-3pm, which would be perfect.

I like driving overnight, but I can't deny that it's very very difficult once the sun comes up.

It would certainly help road safety in meaning that people doing what you suggest wouldn't do it, the car would do it for them. I think the term "very difficult" should be replaced with "highly dangerous". Driving tired is as dangerous as drink-driving, it's only not illegal because it's impossible to quantify it. The only way anyone should drive overnight is if they have properly reset their body clock to a night shift, which for most people takes several days, and nobody should ever drive a 12-13 hour stint without an overnight break - the tachograph laws give a decent indication of what is likely to be safe.

But the idea is that these driverless pods would provide a lifeline to those in rural areas who cannot drive, such as young teenagers and the elderly, in areas where providing a bus service would not be commercially viable. They would be affordable too unlike a taxi which is expensive because it has to cover the cost of the driver.

Rural areas are an interesting question. One of the problems with our ownership-based housing model is that people are reluctant to move. While I get the issue with kids (though parents can give them lifts places, and improving cycling safety would be good) if you're going to give up driving a rural area is really no longer a place for you and moving somewhere more urban, perhaps to an over-55s development, is going to be better.

On the other hand, making rural transport less uneconomic is a potential use-case. There are others, though, such as community bus operations. If you're retired you can plan around fixed day trips e.g. the supermarket is on a Wednesday at 10am every week, coming back at 1pm. You're never, except places served incidentally due to being on a main road or due to high tourist demand, going to find it economic to run hourly services to middle-of-nowhere-land. But single-point DRT (e.g. connecting a set of villages to the nearest railway station for fixed train times) is fairly workable anyway.

 

[Cloud Strife]

It would certainly help road safety in meaning that people doing what you suggest wouldn't do it, the car would do it for them. I think the term "very difficult" should be replaced with "highly dangerous". Driving tired is as dangerous as drink-driving, it's only not illegal because it's impossible to quantify it. The only way anyone should drive overnight is if they have properly reset their body clock to a night shift, which for most people takes several days, and nobody should ever drive a 12-13 hour stint without an overnight break - the tachograph laws give a decent indication of what is likely to be safe.

Yes, there are a lot of problems caused by this in Croatia. People from all over Europe descend on the country on changeover Saturdays, and you always end up with constant crashes on the A1 Zagreb-Ploce motorway. People are driving on Friday evenings to get there for Saturday afternoon, and I've lost count of the number of times that I've seen bad driving from very visibly tired drivers. It's not normally so bad in the return direction, but one of the few roads that frighten me is the section of the A1 between Zagreb and Karlovac on a summer Saturday as people are driving like lunatics just to get to the seaside 5 minutes earlier.

These days, I generally make a point of taking 11 days off work (it's mandatory in PL to take a block of 2 weeks at least once a year), and what I'm doing is driving from Wrocław to Cakovec in Croatia on Friday, then onwards to the coast on Sunday. It makes it a very manageable 8-9 hours driving time on the first day, then 5 hours on the next day. In return, I'm normally leaving on Friday to Cakovec, then Saturday is spent having a very leisurely run home with many breaks after sleeping until 10-11am.

I've done Zadar-Wrocław before in one day, and it was simply dangerous towards the end, even with sharing the driving duties.

This is why I think fully autonomous driving would be a godsend. I wouldn't have to stress about the worst sections of motorway, I would save a huge amount of time, and I'm sure the car will always be a better driver than me after even 4 hours on the road.

In general though, my golden rule is never to drive more than 3 hours without a good break. Unfortunately, my peers in Poland seem to regard it as a badge of honour to drive for hours without stopping. Idiots.

 

[PTR 444]

Rural areas are an interesting question. One of the problems with our ownership-based housing model is that people are reluctant to move. While I get the issue with kids (though parents can give them lifts places, and improving cycling safety would be good) if you're going to give up driving a rural area is really no longer a place for you and moving somewhere more urban, perhaps to an over-55s development, is going to be better.

On the other hand, making rural transport less uneconomic is a potential use-case. There are others, though, such as community bus operations. If you're retired you can plan around fixed day trips e.g. the supermarket is on a Wednesday at 10am every week, coming back at 1pm. You're never, except places served incidentally due to being on a main road or due to high tourist demand, going to find it economic to run hourly services to middle-of-nowhere-land. But single-point DRT (e.g. connecting a set of villages to the nearest railway station for fixed train times) is fairly workable anyway.

In effect, DRT as we know it today will become obsolete when driverless taxis are the norm as why would you wait around for a small minibus to take you round the houses at a fixed time when you could just hail an autonomous car to pick you up anytime without much of an extra cost. You could even utilise sharing of vehicles if for example, several people from multiple villages on a straight line trajectory are headed for the same destination.

 

[Bletchleyite]

In general though, my golden rule is never to drive more than 3 hours without a good break. Unfortunately, my peers in Poland seem to regard it as a badge of honour to drive for hours without stopping. Idiots.

My UK based Polish best mate sees it in this way, too. His driving has calmed down hugely since living in the UK but I still tell him he's a fool for doing long non stop drives like that, which he sometimes does if the flights are a bit pricey.

 

[stuu]

But the idea is that these driverless pods would provide a lifeline to those in rural areas who cannot drive, such as young teenagers and the elderly, in areas where providing a bus service would not be commercially viable. They would be affordable too unlike a taxi which is expensive because it has to cover the cost of the driver.

Unless of course you can automate the bus so enough costs are saved there to make it viable to run a fixed-route, fixed-timetable rural service.

I would have thought that rural areas are the most complicated from the point of view of autonomous vehicles: varying widths, inconsistent lane markings, dodgy geometry, animals with minds of their own etc. A lot of problems to solve, autonomous driving down motorways is orders of magnitude easier, and is pretty much doable now

 

[Bikeman78]

Sure, but things work because generally humans are reasonable people. Some will march out but others will pause a second and let a waiting car through.
Some places this will have to happen or cars waiting to turn will block main roads for ages.

It already happens in the touristy areas of Manhattan. Making a turn is very difficult because there are dozens of pedestrians streaming across until the lights change. Sensible drivers will make as few turns as possible. I've never driven in the US and managed to work that out.