How could self driving vehicles effect future rail growth?

[Technologist]

I have said many times on simular threads.

True self driving cars will cause gridlock. Think about a family of 4. 2 x adults with a car then their 10 and 12 year old children. Currently there will be one or two cars on the drive. True driverless cars you will mean 4 cars, one each. And when the kids go around to their mates birthday party and 10 driverless cars all turn up on a street to park immagine the chaos. Immagine Schools with all the cars.

If driverless cars go down the Uber route of being taxi's then every journey will result in multiple more road journeys. Vehicle drives to pick your up, drives you tl your destination then goes to another pick up its next job so 2/3 journeys rather than one.

Grislock is my prediction or very slow journeys.

Raikways, especialy urban ones will be very busy as it wil be so much quicker to travel.

The grid lock stage will then give political cover to implement demand based road charging which will then cause people share journeys when they go through a choke point. Ride sharing is pretty easy to implement through a self driving Uber type system and the delays picking people up will be vastly less than the time saving be removing all congestion.

With a bit of luck the fear of grid lock will allow us to implement the road charging before it actually happens!

At the moment ride sharing in taxis is pretty uncommon, this is because taxis are relatively uncommon so you would be waiting a while for one to arrive and the chances of finding someone going on a similar journey is low. If literally every car is a taxi pairing people up is pretty easy. Instead of one car per child its more like 8-10 kids in an autonomous minibus. For trips where people don't know each other I suspect that robotaxis will be designed so you can be completely separated from other passengers if you want.

To put this into context cars do about 10X the passenger miles that trains manage and do so with about and average of 1.6 people on board, ergo you could add the entire impact of the rail network by adding an addition 0.16 people per car without adding any cars to the network.

This is before we realise that self driving cars are to cars what "horseless carriages" are to carriages. Your car weighs two tonnes because it's be sized around a yearly holiday where all the family go and bring bikes and a dog, most of the time it's just got one person in it. It is likely that the default robotaxi will have two seats, when you go on holiday you'd just hire an 8 seater or just hire multiple vehicles, or have Amazon just take you luggage to the hotel. If your journey never exits and urban area then it's likely that your robotaxi will be relatively short ranged and not particularly high performance.

Thus the most common self driving vehicle might be a self balancing enclosed motorbike (see Ecomobile) or a leaning tri-cycle, both vehicles that can ride two up in a lane so we have just doubled road space and that's before we include capacity increases due to robots interacting more predictably with each other. We could get rid of non self driving cars surprisingly quickly by just banning them from any remotely high traffic area, it's relatively easy to add self driving to any post 2010 car particularly if it's automatic and if you have people who are hold outs in the world where ubiquitous self driving vehicles exist it's pretty easy to just summon a self driving recovery platform to carry any old non-self driving car across a restricted area (the future equivalent of a horse box).

 

[Bikeman78]

I have said many times on simular threads.

True self driving cars will cause gridlock. Think about a family of 4. 2 x adults with a car then their 10 and 12 year old children. Currently there will be one or two cars on the drive. True driverless cars you will mean 4 cars, one each. And when the kids go around to their mates birthday party and 10 driverless cars all turn up on a street to park immagine the chaos. Immagine Schools with all the cars.

If driverless cars go down the Uber route of being taxi's then every journey will result in multiple more road journeys. Vehicle drives to pick your up, drives you tl your destination then goes to another pick up its next job so 2/3 journeys rather than one.

Grislock is my prediction or very slow journeys.

Raikways, especialy urban ones will be very busy as it wil be so much quicker to travel.

There are so many unknown variables. I'm not sure that having lots of cars driving around empty is winning at life. Won't that increase the amount of traffic? Road markings and signs would need to be maintained properly, rather than being worn away. Will they mount the pavement on narrow streets like current cars do? Imagine the fallout when one of them kills someone. If one of these cars runs down a pedestrian or cyclist, who will be liable?

 

[bspahh]

True self driving cars will cause gridlock. Think about a family of 4. 2 x adults with a car then their 10 and 12 year old children. Currently there will be one or two cars on the drive. True driverless cars you will mean 4 cars, one each. And when the kids go around to their mates birthday party and 10 driverless cars all turn up on a street to park imagine the chaos. Imagine Schools with all the cars.

When technology changes, then the use patterns will change. If you have your own driverless car, then it still needs to park somewhere when you aren't using it. However, that parking area does not have to be at your home or your destination. You could get the car drive to somewhere with cheap parking, and then come and pick you up when you are done.

It then becomes a pricing issue. Surge pricing for road use can manage congestion. Humans don't get on well with booking a timed slot to use a road, but thats the sort of thing which is fine for a self-driving car. There are times when you would want to journeys to be at a specific time such as for an appointment. There are lots of times when "sometime this afternoon" would be fine.

If driverless cars go down the Uber route of being taxi's then every journey will result in multiple more road journeys. Vehicle drives to pick your up, drives you to your destination then goes to another pick up its next job so 2/3 journeys rather than one.

A automated taxi means you get dropped off near your destination, rather than having to find a parking space. It would also mean that kids could be dropped off at social events and activities without their parent having to then hang around until they are done.

With your family of 4 with 2 kids, that has 2 cars at the moment, I think you might have one car that is big enough for everyone, for trips when public transport isn't running. A second car would be replaced with an account for self-driving taxis, and perhaps a small self-driving car for local journeys.

 

[stevieinselby]

With your family of 4 with 2 kids, that has 2 cars at the moment, I think you might have one car that is big enough for everyone, for trips when public transport isn't running. A second car would be replaced with an account for self-driving taxis, and perhaps a small self-driving car for local journeys.

One of the determining factors will be whether the robotaxi companies flood the market or throttle demand.
It will be the same thing we've seen with DRT – you can have massive oversupply, meaning that anyone can travel whenever they want, but most of the time they're sitting idle, which is not compatible with a commercially run system – or you ration supply to the extent that most cars will be in use for most of the time, so that they are nicely profitable, but this means that people often won't be able to travel when they want to, certainly not without booking in advance, and that isn't going to drive any kind of modal shift because it isn't convenient.
If households haven't given up their second car for public transport or taxis then it's unlikely that robotaxis will offer anything sufficiently different for most of them to make the switch.

 

[Bletchleyite]

It then becomes a pricing issue. Surge pricing for road use can manage congestion. Humans don't get on well with booking a timed slot to use a road, but thats the sort of thing which is fine for a self-driving car. There are times when you would want to journeys to be at a specific time such as for an appointment. There are lots of times when "sometime this afternoon" would be fine.

I don't think people would, with cars, be willing to book a time in advance, unless the discount was massive. A model more like Uber where the price now is influenced by demand on a far more dynamic basis than Advance tickets, on the other hand, may work.

To be honest I'm astonished it hasn't been applied to the M6 Toll. When it's heaving you could easily charge more, but when it's really quiet people might go on it for a few quid who wouldn't for a tenner.

 

[Krokodil]

I'm not sure that having lots of cars driving around empty is winning at life. Won't that increase the amount of traffic?

Uber certainly increased congestion for the same reason.

 

[Bletchleyite]

Uber certainly increased congestion for the same reason.

A slightly different one. Cheaper and easier to get taxis = more taxis. People switched from public transport.

 

[Krokodil]

A slightly different one. Cheaper and easier to get taxis = more taxis. People switched from public transport.

But also more deadheading increasing congestion

 

[Bletchleyite]

But also more deadheading increasing congestion

Compared to driving yourself, yes. But not that many people drive themselves in central London, which is where this was primarily observed. They switched from buses and the Tube because an Uber was easy to get and cost, per head, less than public transport.

 

[Sorcerer]

Compared to driving yourself, yes. But not that many people drive themselves in central London, which is where this was primarily observed. They switched from buses and the Tube because an Uber was easy to get and cost, per head, less than public transport.

Given that Uber charges more when demand is higher, this might be one of those instances where the problem of potential congestion actually solves itself because the cost for public transport will become cheaper before enough people get enough Ubers to cause traffic jams, therefore encouraging them to use public transport and reduce road traffic beyond those prepared to pay for it. On a national scale though I can't see it working nearly as well.