Could railways become high speed roads for automated vehicles?

[Wynd]

No. This is a ridiculus, impractical, and non-sane approach.

Autonomous vehicles create many of the same problems we already have.

 

[coppercapped]

It wouldn't be a death trap though, any more than standing by the platform at Northallerton as an Azuma goes through at 125mph inches away is today. Be safer if anything. The super brains of the cars would be able to avoid you (and if unsure about that would already be going slower).

Please don't get carried away with predicting the future based on the concept of cars' 'super brains', they are further in the future than is commonly assumed.

The (American) Institute of Electrical and Electronic Engineers (the IEEE) publishes Spectrum, a monthly journal containing articles of general interest in a variety of technical areas for engineers to inform them of developments in fields other than their own. The current issue contains a summary of the present state of the art for robot/self-driving cars (M. L. Cummings, IEEE Spectrum, Vol. 60. issue 10, October 2023, pp30-5) and explains many of their limitations. I quote:

The lack of technical comprehension across industry and government is appalling. People do not understand that the AI (Artificial Intelligence) that runs vehicles — both the cars that operate in self-driving modes and the much larger number of cars offering advanced driving-assistance systems (ADAS) — are based on the same principles as ChatGPT and many other large language models (LLMs). These systems control a car's lateral and longitudinal position — to change lanes, brake and accelerate — without waiting for orders to come from the person sitting behind the wheel.
Both kinds of AI use statistical reasoning to guess what the next word or phrase or steering input should be, heavily weighting the calculations with recently used words or actions. Go to your Google search window and type in "now is the time" and you will get the result "now is the time for all good men". And when your car detects an object on the road ahead, even if it's just a shadow, watch the car's self-driving module suddenly brake.
Neither the AI in LLMs nor the one in autonomous cars can "understand" the situation, the context, or any unobserved factors that a person would consider in a similar situation. The difference is that while a language model may give you nonsense, a self-driving car can kill you.

The author then lists five factors that need to be considered in the design and use of AI in self-driving vehicles:

1. Human errors in (vehicle) operation get replaced by human errors in coding
2. AI failure modes are hard to predict
3. Probabilistic estimates do not approximate to judgement under uncertainty
4. Maintaining AI is just as important as creating AI
5. AI has system-level implications that can't be ignored.

Each of these factors is discussed in depth but I find the last point interesting in view of comments and statements made earlier in this thread. The author writes:

Self-driving cars have been designed to stop cold the moment they can no longer reason and no longer resolve uncertainty. This is an important safety feature. But as Cruise, Tesla and Waymo have demonstrated, managing such stops poses and unexpected challenge.
A stopped car can block roads and intersections, sometimes for hours, throttling traffic and keeping out first-response vehicles. Companies have instituted remote-monitoring centers and rapid-action teams to mitigate such congestion and confusion, buheir respt at least in San Francisco, where hundreds of self-driving cars are on the road, city officials have questioned the quality of their responses.
Self-driving cars rely on wireless connectivity to maintain their road awareness, but what happens when that connectivity drops? One driver found out the hard way when his car became entrapped in a knot of 20 Cruise vehicles that had lost connection with the remote operations center and caused massive traffic jams.

The author does not call for a ban on autonomous vehicles but does call for much more oversight on the use of AI in this application. I would conclude that the wide scale use of self-driving vehicles on normal roads is many years further in the future than is commonly supposed.

 

[Sorcerer]

I think rail outside of The Underground will cease to exist when road traffic automation has been adopted as the rails can be removed and become thoroughfares for high speed automated road vehicles. They are perfect for this type of evolution.

I suppose automatic road transport will certainly prove itself to be better and more efficient than railways, therefore I think we should rip up all railways outside the capital to pave way for new motorways of autonomous vehicles. We can even futureproof it slightly by increasing capacity with autonomous buses so we don't cram these roads up too much. In fact for anyone who doesn't want to or cannot drive, we should run several of these buses several times per day to transport large numbers of people. To increase capacity further, should build dedicated busways for these buses, and maybe we should even couple a few of these buses together so there is one centralised computer controlling the entire trail of buses behind it rather than having lots of autonomous individual buses transporting less people.

But then again that's a lot of weight on the rubber tyres that will generate a lot of friction between the wheels and tarmac and lead to lots of wear down, but I have a solution for this. We should replace the bus tyres and wheels with single steel wheels to run along guided rails to reduce rolling resistance for more energy efficient movement. Since we're running it along a guided rail we might as well also add some electric wires for the bus to collect electricity to power the engines instead of batteries or diesel/petrol engines which can save on energy costs long-term as well allowing us to increase speeds and fit more of these super long buses on the roadways. This new mode of transport consisting of electric coaches all coupled together powered by a single cab at the front running along guided rails shall be named... nope, sorry, my mistake, I just turned the cars into buses and the buses into trains.

The ultimate point of my long-winded scenario here is to demonstrate that replacing railways with high-speed automated roadways and road vehicles is not actually an evolution like you say it is, but it's actually a devolution that will be worse in almost every possible way. The idea you have presented here sounds remarkably similar to The Boring Company's Las Vegas Loop which combines all of the expenses of a heavy rail metro system with the disadvantages of road traffic without the advantages of either of them. Roadways with automatic cars will be slower, less energy efficient, and have less capacity than a railway when it comes to mass transport, therefore it would make absolutely no sense to replace them with automatic roadways like you're suggesting, which is why despite our current motorway infrastructure filling it's own niche, we still have railways filling theirs.

 

[Egg Centric]

Please don't get carried away with predicting the future based on the concept of cars' 'super brains', they are further in the future than is commonly assumed.

The (American) Institute of Electrical and Electronic Engineers (the IEEE) publishes Spectrum, a monthly journal containing articles of general interest in a variety of technical areas for engineers to inform them of developments in fields other than their own. The current issue contains a summary of the present state of the art for robot/self-driving cars (M. L. Cummings, IEEE Spectrum, Vol. 60. issue 10, October 2023, pp30-5) and explains many of their limitations. I quote:


The author then lists five factors that need to be considered in the design and use of AI in self-driving vehicles:

1. Human errors in (vehicle) operation get replaced by human errors in coding
2. AI failure modes are hard to predict
3. Probabilistic estimates do not approximate to judgement under uncertainty
4. Maintaining AI is just as important as creating AI
5. AI has system-level implications that can't be ignored.

Each of these factors is discussed in depth but I find the last point interesting in view of comments and statements made earlier in this thread. The author writes:


The author does not call for a ban on autonomous vehicles but does call for much more oversight on the use of AI in this application. I would conclude that the wide scale use of self-driving vehicles on normal roads is many years further in the future than is commonly supposed.

I wasn't necessarily expecting this in even either Rob or I's lifetime, never mind next year. It just seems a reasonable destination to me on current trends with no grand physical reason (I'm aware of) it cannot happen.

 

[6Gman]

Don't forget that in the future autonomous cars would be able to travel far quicker than current human driven cars due to what will eventually be their instantaneous reactions and ability to see round bends / coordinate with each other due to *waves hands* some kind of networking. It's not therefore clear that roads will have to be upgraded.

The reaction may be instantaneous but they still need a braking distance given the need to protect the occupants of the vehicle.

60 to 0 in half a second doesn't do a lot for the internal organs.

 

[Ashfordian6]

I suppose automatic road transport will certainly prove itself to be better and more efficient than railways, therefore I think we should rip up all railways outside the capital to pave way for new motorways of autonomous vehicles. We can even futureproof it slightly by increasing capacity with autonomous buses so we don't cram these roads up too much. In fact for anyone who doesn't want to or cannot drive, we should run several of these buses several times per day to transport large numbers of people. To increase capacity further, should build dedicated busways for these buses, and maybe we should even couple a few of these buses together so there is one centralised computer controlling the entire trail of buses behind it rather than having lots of autonomous individual buses transporting less people.

But then again that's a lot of weight on the rubber tyres that will generate a lot of friction between the wheels and tarmac and lead to lots of wear down, but I have a solution for this. We should replace the bus tyres and wheels with single steel wheels to run along guided rails to reduce rolling resistance for more energy efficient movement. Since we're running it along a guided rail we might as well also add some electric wires for the bus to collect electricity to power the engines instead of batteries or diesel/petrol engines which can save on energy costs long-term as well allowing us to increase speeds and fit more of these super long buses on the roadways. This new mode of transport consisting of electric coaches all coupled together powered by a single cab at the front running along guided rails shall be named... nope, sorry, my mistake, I just turned the cars into buses and the buses into trains.

The ultimate point of my long-winded scenario here is to demonstrate that replacing railways with high-speed automated roadways and road vehicles is not actually an evolution like you say it is, but it's actually a devolution that will be worse in almost every possible way. The idea you have presented here sounds remarkably similar to The Boring Company's Las Vegas Loop which combines all of the expenses of a heavy rail metro system with the disadvantages of road traffic without the advantages of either of them. Roadways with automatic cars will be slower, less energy efficient, and have less capacity than a railway when it comes to mass transport, therefore it would make absolutely no sense to replace them with automatic roadways like you're suggesting, which is why despite our current motorway infrastructure filling it's own niche, we still have railways filling theirs.

Rail and public transport usage will reduce once autonomous vehicles are in use because they offer point to point convenience that something on guided rails can never offer. It can travel at your time and will not need to stop until it reaches its destination.

With the increase in individual travel via autonomous vehicles, the cost of running railways will become prohibitive as usage drastically drops. Routes will then start to be converted into dedicated high speed thoroughfares for these autonomous vehicles.

I suspect all London terminus stations will become dead-end drop off areas for autonomous vehicles and the vehicle you are in will be able to join the network at dedicated points on the route. The car will then disappear onto its next job. Despite not taking you to your exact location, they will be highly used because of their high speed access to central London swiftly because they will not be slowed down by interacting with the current road network, or slowed by pedestrians and cyclists.

Efficiency will not need to be a consideration as they will be electric vehicles powered by renewable energy.

P.S. You will not own a autonomous car, it will be some Uber style on-demand service as these services will massive undercut car ownership.

 

[The exile]

I could potentially see it happening for lightly used regional branch lines such as Plymouth - Gunnislake and the Conway Valley Line, but definitely not on any busy commuter or intercity routes.

Except of course many of these were built as single track…

P.S. You will not own an autonomous car, it will be some Uber style on-demand service as these services will massive undercut car ownership.

So what happens to the child seats or the bulky things that you are going to deliver on the way home. It all sounds so perfect on paper (just as the “freedom of the open road’ did 70-odd years ago) but needs to confront realities of how people actually use cars.

 

[Krokodil]

If all cars are autonomous I would just walk out into the road when I want to cross knowing that the car will be forced to stop.

Imagine the fun you could have by walking out into the middle of the road with a traffic cone, placing it down and walking off to watch everything snarl up woth everyone locked into their autonomous taxis. Much less effort than digging a trench across Piccadilly.

Horace de Vere Cole - Wikipedia

en.wikipedia.org

Indeed. We'll all be forced to make ridiculous detours so as not to get in the way of the all important cars.

Nothing new there then. Shame that we've just got to the point of giving the streets back to the people.

Rail and public transport usage will reduce once autonomous vehicles are in use because they offer point to point convenience that something on guided rails can never offer. It can travel at your time and will not need to stop until it reaches its destination.

They'll still get caught in traffic jams (even with no pranksters in sight).

 

[Sorcerer]

Rail and public transport usage will reduce once autonomous vehicles are in use because they offer point to point convenience that something on guided rails can never offer. It can travel at your time and will not need to stop until it reaches its destination.

With the increase in individual travel via autonomous vehicles, the cost of running railways will become prohibitive as usage drastically drops. Routes will then start to be converted into dedicated high speed thoroughfares for these autonomous vehicles.

Tell me, if regular cars and a vast motorway network for the past sixty years haven't been able to render rail transport obsolete, what makes you think that cars that drive themselves but are otherwise identical will somehow be able to do so?

I suspect all London terminus stations will become dead-end drop off areas for autonomous vehicles and the vehicle you are in will be able to join the network at dedicated points on the route. The car will then disappear onto its next job. Despite not taking you to your exact location, they will be highly used because of their high speed access to central London swiftly because they will not be slowed down by interacting with the current road network, or slowed by pedestrians and cyclists.

If you will have a network of cars that will travel from dedicated destinations at dedicated pick-up and drop-off points why even bother replacing the railways to begin with when the added flexibility of being able to hire it when you want will only be offset by the fact it will be unable to carry as many people and luggage as well as being caught up in traffic that may just end up making your journey longer?

Efficiency will not need to be a consideration as they will be electric vehicles powered by renewable energy.

Energy efficiency is not the only efficiency at play here. Cars won't be able to carry as many people or luggage as trains and will be bound by the same stopping and braking distances as trains assuming they would even be able to go as fast, along with the added problem of other traffic on the dedicated roadway, but they will also have the disadvantage of greater wear and tear because rubber tyres on asphalt or tarmac road surfaces has greater rolling resistance than steel wheels on steel rails. Therefore it will proportionally take more energy to move these autonomous vehicles along the road than it would for a train.

P.S. You will not own a autonomous car, it will be some Uber style on-demand service as these services will massive undercut car ownership.

Thank you for basically confirming that your idea is merely a scaled up version of the Las Vegas Loop which combines all of the expenses of a railway system with the disadvantages of road traffic without enjoying any of the benefits of either of them.

 

[deltic]

Rail and public transport usage will reduce once autonomous vehicles are in use because they offer point to point convenience that something on guided rails can never offer. It can travel at your time and will not need to stop until it reaches its destination.

With the increase in individual travel via autonomous vehicles, the cost of running railways will become prohibitive as usage drastically drops. Routes will then start to be converted into dedicated high speed thoroughfares for these autonomous vehicles.

I suspect all London terminus stations will become dead-end drop off areas for autonomous vehicles and the vehicle you are in will be able to join the network at dedicated points on the route. The car will then disappear onto its next job. Despite not taking you to your exact location, they will be highly used because of their high speed access to central London swiftly because they will not be slowed down by interacting with the current road network, or slowed by pedestrians and cyclists.

Efficiency will not need to be a consideration as they will be electric vehicles powered by renewable energy.

P.S. You will not own a autonomous car, it will be some Uber style on-demand service as these services will massive undercut car ownership.

Public transport is loss making because its peak vehicle requirement is far higher than its off-peak requirement. A private sector uber type service will, as now, have high surge prices at peak demand but even higher. Raising huge social issues. Uber cars have an average of 40% of empty running - replicated by all cars will result in far more traffic on the roads in major towns/cities. If, and its a big if, autonomous vehicles work as promised I can see them becoming the new rural public transport service. However, the chances of replacing all private cars is a non-starter.

 

[Wynd]

Rail and public transport usage will reduce once autonomous vehicles are in use because they offer point to point convenience that something on guided rails can never offer. It can travel at your time and will not need to stop until it reaches its destination.

With the increase in individual travel via autonomous vehicles, the cost of running railways will become prohibitive as usage drastically drops. Routes will then start to be converted into dedicated high speed thoroughfares for these autonomous vehicles.

I suspect all London terminus stations will become dead-end drop off areas for autonomous vehicles and the vehicle you are in will be able to join the network at dedicated points on the route. The car will then disappear onto its next job. Despite not taking you to your exact location, they will be highly used because of their high speed access to central London swiftly because they will not be slowed down by interacting with the current road network, or slowed by pedestrians and cyclists.

Efficiency will not need to be a consideration as they will be electric vehicles powered by renewable energy.

P.S. You will not own a autonomous car, it will be some Uber style on-demand service as these services will massive undercut car ownership.

This cant actually be a serious post, can it?

How far in to the future are you thinking here? You are aware that the capital costs of what you are proposing are gargantuan. Who is paying for it?

Efficiency not a consideration? Are you serious?

You say point to point conveneince then say they wont take you to your exact destination? Which one is it?

The more you think about this, the more banans it gets.

Imagine Reading to Paddington? Or KGX to York. Or Crossrail. Its not even the beginning of a good idea, let alone a sane and ratrional one.

 

[Egg Centric]

The reaction may be instantaneous but they still need a braking distance given the need to protect the occupants of the vehicle.

60 to 0 in half a second doesn't do a lot for the internal organs.

I envisage a sliding scale where you can configure your car per personal preference, starting at "sedate", passing through "sensible" to "sporting" and finally "lunatic". The latter unlocks a 2/3G mode with burst to 5 or 6G.

Clearly the speed still has some sort of limit but you could achieve a lot with that (remember they can swerve as well).

This also reminds me of the genius of my roller coaster train which remains a great proposal far ahead of its time imo.

 

[Merle Haggard]

An reverse scenario would be to close Motorways and lay railway tracks on them.
Think how easily HS2 and its extensions could be built using the alignment of the M1 to Catthorpe then the M6.

 

[Krokodil]

In the US they are using the central reservation of their highways to build rail lines. Their highways are pretty long and straight though

 

[yorkie]

Rail and public transport usage will reduce once autonomous vehicles are in use because they offer point to point convenience that something on guided rails can never offer. It can travel at your time and will not need to stop until it reaches its destination.

With the increase in individual travel via autonomous vehicles, the cost of running railways will become prohibitive as usage drastically drops. Routes will then start to be converted into dedicated high speed thoroughfares for these autonomous vehicles.

I suspect all London terminus stations will become dead-end drop off areas for autonomous vehicles and the vehicle you are in will be able to join the network at dedicated points on the route. The car will then disappear onto its next job. Despite not taking you to your exact location, they will be highly used because of their high speed access to central London swiftly because they will not be slowed down by interacting with the current road network, or slowed by pedestrians and cyclists.

Efficiency will not need to be a consideration as they will be electric vehicles powered by renewable energy.

P.S. You will not own a autonomous car, it will be some Uber style on-demand service as these services will massive undercut car ownership.

What year are you predicting this for? I'm tempted to set a calendar reminder with the URL to this thread!

 

[Bald Rick]

In the US they are using the central reservation of their highways to build rail lines. Their highways are pretty long and straight though

And have big central reservations! Try that on the M1 and it would need to be 7.25” gauge!

More seriously, not many railways are being built in the “median strip”, and notably the new Brightline railway from Cocoa to MCO Orlando Airport is specifically one side or the other. The Washington Metro has a couple of examples in the median including a section of the Silver line in the middle of the Dulles Access Road, and then LA has one Metro line in the middle of the Century Freeway - which makes it spectacularly difficult to get to for passengers; look at Harbor Freeway station as an example.

 

[Krokodil]

More seriously, not many railways are being built in the “median strip”, and notably the new Brightline railway from Cocoa to MCO Orlando Airport is specifically one side or the other. The Washington Metro has a couple of examples in the median including a section of the Silver line in the middle of the Dulles Access Road, and then LA has one Metro line in the middle of the Century Freeway - which makes it spectacularly difficult to get to for passengers; look at Harbor Freeway station as an example.

At no point was I suggesting that it was a good idea. It's as much a good idea as the OP's. There aren't many good ideas originating from the US, especially where transport is concerned.

I'll add a few examples to your list, such as the Chicago Blue Line, though they were built simultaneously rather than as an add-on. Brightline West is proposed to go down the middle of Interstate 15 (incidentally this project appears to be a decent piece of HSR as opposed to the bizarre and frankly dangerous setup in Florida).

 

[Bald Rick]

At no point was I suggesting that it was a good idea. It's as much a good idea as the OP's. There aren't many good ideas originating from the US, especially where transport is concerned.

I'll add a few examples to your list, such as the Chicago Blue Line, though they were built simultaneously rather than as an add-on. Brightline West is proposed to go down the middle of Interstate 15 (incidentally this project appears to be a decent piece of HSR as opposed to the bizarre and frankly dangerous setup in Florida).

You missed the air train which technically is routed down the median strip of the Van Wyck Expressway to JFK, if mostly about 50ft above it!

 

[Clayton]

I suppose automatic road transport will certainly prove itself to be better and more efficient than railways, therefore I think we should rip up all railways outside the capital to pave way for new motorways of autonomous vehicles. We can even futureproof it slightly by increasing capacity with autonomous buses so we don't cram these roads up too much. In fact for anyone who doesn't want to or cannot drive, we should run several of these buses several times per day to transport large numbers of people. To increase capacity further, should build dedicated busways for these buses, and maybe we should even couple a few of these buses together so there is one centralised computer controlling the entire trail of buses behind it rather than having lots of autonomous individual buses transporting less people.

But then again that's a lot of weight on the rubber tyres that will generate a lot of friction between the wheels and tarmac and lead to lots of wear down, but I have a solution for this. We should replace the bus tyres and wheels with single steel wheels to run along guided rails to reduce rolling resistance for more energy efficient movement. Since we're running it along a guided rail we might as well also add some electric wires for the bus to collect electricity to power the engines instead of batteries or diesel/petrol engines which can save on energy costs long-term as well allowing us to increase speeds and fit more of these super long buses on the roadways. This new mode of transport consisting of electric coaches all coupled together powered by a single cab at the front running along guided rails shall be named... nope, sorry, my mistake, I just turned the cars into buses and the buses into trains.

The ultimate point of my long-winded scenario here is to demonstrate that replacing railways with high-speed automated roadways and road vehicles is not actually an evolution like you say it is, but it's actually a devolution that will be worse in almost every possible way. The idea you have presented here sounds remarkably similar to The Boring Company's Las Vegas Loop which combines all of the expenses of a heavy rail metro system with the disadvantages of road traffic without the advantages of either of them. Roadways with automatic cars will be slower, less energy efficient, and have less capacity than a railway when it comes to mass transport, therefore it would make absolutely no sense to replace them with automatic roadways like you're suggesting, which is why despite our current motorway infrastructure filling it's own niche, we still have railways filling theirs.

Yes good one. This scenario comes around every now and then and this is a good riposte. Autonomous cars will still get in the way and be ‘traffic’.

 

[Clayton]

Also, the idea of using hired autonomous vehicles falls down once you think of the reality. They’d be full of crisp packets and burger boxes and someone’s done a wee in the corner. How often would they need to return to base for cleaning? And many people would be too self-important to use a shared vehicle.

 

[yorksrob]

Also, the idea of using hired autonomous vehicles falls down once you think of the reality. They’d be full of crisp packets and burger boxes and someone’s done a wee in the corner. How often would they need to return to base for cleaning? And many people would be too self-important to use a shared vehicle.

Indeed. The whole USP of motor transport is having your own bubble on wheels.

 

[778]

Also, the idea of using hired autonomous vehicles falls down once you think of the reality. They’d be full of crisp packets and burger boxes and someone’s done a wee in the corner. How often would they need to return to base for cleaning? And many people would be too self-important to use a shared vehicle.

I can imagine them being full of sick on Friday and Saturday nights. Would a young woman want to get into an autonomous vehicle with a group of drunk men?

This could increase the carrying capacity of that network negating the need for costly new roads. However, autonomous vehicles are more likely to struggle in urban areas due to the interaction with pedestrians and cyclists.

I would have thought rural roads (narrow country lanes) will present more of a challenge for Autonomous vehicles? I think the first place you will see them will be in city centres.

 

[PTR 444]

I would have thought rural roads (narrow country lanes) will present more of a challenge for Autonomous vehicles? I think the first place you will see them will be in city centres.

Personally, I believe autonomous vehicles will be trialled in self-contained areas with a very low traffic and population density, such as Orkney, Shetland and the Western Isles, before being rolled out elsewhere.

 

[deltic]

Personally, I believe autonomous vehicles will be trialled in self-contained areas with a very low traffic and population density, such as Orkney, Shetland and the Western Isles, before being rolled out elsewhere.

They are already being trialled in the UK in urban areas eg Woolwich London earlier this year and Stagecoach ran its autonomous bus for around 6 months this year between Inverkeithing and Edinburgh. It is apparently going through an upgrade at the moment and is back with a human driver. Milton Keynes is presently hosting a trial now of a shuttle bus as this BBC article explains https://www.bbc.co.uk/news/uk-england-beds-bucks-herts-67418123

.The vehicle will be on selected roads throughout November as part of a Europe-wide research project.

 

[Technologist]

No, for several reasons.
1). Even with coach type vehicles, the surface would be impacted and need resurfacing periodically (frequent heavier vehicles lead to swifter road surface deterioration), plus you'd still need frequent tyre replacement and have particulate pollution from tyre particles, unless you retained the rails!
2) A vehicle on rails is one of the most efficient forms of transport, due to the reduction of friction at the interaction with the surface.
Per passenger, the road vehicle will need more energy to power it at the same speed.
3) Tarmac surfacing cannot handle the same range of temperatures as rails can.
E.g. during wet weather, the railway itself can typically perform as normal (issues often result from items external to the railway falling onto the railway during a storm).
As for road traffic, speeds have to be reduced, accidents are more likely to happen etc.

1: Road wear is proportional to the 4th power of axle load, roadware is this disproportionately due to HGVs which we wouldn't have on a high speed autonomous vehicle "line". Tyres are not a big cost of motoring and roads are much more tolerant of wear than rails ergo on a per passenger mile basis we spend much less on road maintenance than rail maintenance.

2: Friction with the road surface is a relatively low proportion of the energy expanded moving passengers. If you get medium capacity autonomous road vehicles to platoon you functionally have a train, but with much more flexibility and lower infrastructure costs.

3: Even in the pissing wet rubber on tarmac will outbrake a steel on steel interface. Weather causes accidents because humans are poor judges and behave inconsistently. Autonomous vehicles driving on grade separated roads will be quite capable of driving at high speeds and with reduced headway compared to human drivers.

Autonomous electric cars are very rapidly going to diverge from the form factors and usage patterns of current road vehicles and will become something quite different.

 

[yorksrob]

It would be nice if someone invented something useful, like nuclear fusion power.

Instead we get more cars and AI.

What an age we live in.