Automating the roads - (split from Flying Cars thread)

[SH3RL0CK]

Faster, more personal freedom, etc.

 

On a personal note, I live just a short distance (easy with a bike) and I work about a block from a train station in a major U.S. city (Chicago) which incidentally has great public transportation (I would judge it to be comparable to the European and Japanese cities I have visited). However I still chose to drive to work (about 45 minutes each way) because the equivalent train commute would take around 3 hours each way (1 transfer) as the system will not take me directly. With several bus and train transfers I could potentially cut this down to about 2.5 hours . So while I would like to take the train, its very inconvenient and I'm not planing on doing so.

[CaptainPanic]

hmm... that's a rather huge difference. (Much larger than I'd expect). Do you drive so fast, or is the train so slow, or the connection/transfer so bad? Or do you have to walk so far on the other end of the line?

 

In Europe, intercity and international trains should average about 80 km/h, I think (including stops and transfers), so you're not so much slower than driving, especially if you include a coffee break, fueling and possible traffic jams. And then there are the high speed trains which average at 250 km/h, and have top speeds well over 300 km/h. The high speed network is still expanding.

 

Our public transportation only sucks inside cities (bus systems are a disaster, trams are slow and only the metro does its job sometimes, if you're lucky to live in a place where they have one). Then again, the bus is for the elderly who cannot walk so far, can no longer drive and have lots of time

[SH3RL0CK]

I don't really drive fast and the trains are comparable to the car speed. The difference is that I work and live on the outskirts of the metro area and the train lines only meet in the central "hub". As such, to take the train I have to go all the way in, then all the way back out as opposed to a direct path. The alternatives to connect while still on the outside portion of the metro area are taking multiple buses and/or walking which adds to the time required. I just don't have the option to take the hypotenuse of the triangle.

 

The other problem is that one of the lines in particular has many stops, which adds to the time required.

[Sisyphus]

In response to specific objections about speed/fossil fuel use.

 

First off, I absolutely stand by the "approaching 300mph" claim. France's TGV recently broke it's own record and hit a top speed of 357mph. The maglev train in Shanghai that runs from the airport to the downtown area (about 20 miles) hits a top speed of about 270mph on every run.

 

As for "still using fossil fuels," it's far easier to make trains electric (many actually are, including some of the fastest and most used) than it is for cars. Their range is unlimited, they don't have to carry their fuel with them, and the infrastructure is built in.

[npts2020]

Sisyphus; And the Germans have tested their trains up to 581 kph but America has no similar high speed trains. Very few freight trains are electric either and if you are going to build that kind of infrastructure (electric) anyway, why build it for huge trains that many people will not use for the same reasons SH3RLOCK has articulated. High-speed/heavy rail is the most expensive infrastructure we could possibly build and the least flexible for getting where you want to go. IMO one of the biggest technical challenges for automation is what CaptainPanic has stated about intersections. The problem there is that roadways are based on a centuries old design that predates the Romans "paving" the Appian Way with flagstones, and were never meant for high-speed traffic. Even an automated system will not achieve high speeds in an urban setting where there are pedestrians, bicyclists and the like competing for the same space. Intersections are the point where all of these competitors get in each other's way. While difficult, the problem is not intractable. One way of going about it would be to elevate one direction (N/S or E/W) and have the other run at street level in heavily populated areas. Another might be to redesign some intersections so that they are like a bridge where crosstraffic passes under or over. It would not be necessary to have as many cross streets as there are now for the same reasons every street doesn't cross a river (or even small creek).

[CaptainPanic]

Sisyphus; And the Germans have tested their trains up to 581 kph but America has no similar high speed trains. Very few freight trains are electric either and if you are going to build that kind of infrastructure (electric) anyway, why build it for huge trains that many people will not use for the same reasons SH3RLOCK has articulated.

 

npts2020, can you provide a link to the claim that the Germans have tested their trains at such incredible speeds? And I'd like to say that many of our freight trains are electric. There exist also companies that operate on lines that have no electric infrastructure, so they use diesel locomotives.

 

I also don't see your point of "why build it for huge trains that many people will not use"? Train lines between large cities have intercity trains every 15 minutes, and less in the night. These trains are often half full, and totally packed in the rush hours. That means that up to 10000 people travel on one line per day, which is in the same order of magnitude as highways... and the investment to build either a highway or a trainline is the same order of magnitude as well.

 

High-speed/heavy rail is the most expensive infrastructure we could possibly build and the least flexible for getting where you want to go. IMO one of the biggest technical challenges for automation is what CaptainPanic has stated about intersections. The problem there is that roadways are based on a centuries old design that predates the Romans "paving" the Appian Way with flagstones, and were never meant for high-speed traffic. Even an automated system will not achieve high speeds in an urban setting where there are pedestrians, bicyclists and the like competing for the same space. Intersections are the point where all of these competitors get in each other's way. While difficult, the problem is not intractable. One way of going about it would be to elevate one direction (N/S or E/W) and have the other run at street level in heavily populated areas. Another might be to redesign some intersections so that they are like a bridge where crosstraffic passes under or over. It would not be necessary to have as many cross streets as there are now for the same reasons every street doesn't cross a river (or even small creek).

 

Investments for a large amount of bridges might be just as high as a bunch of railroads. In both cases you design an infrastructure where you want to keep several flows of traffic separated (in one case, you separate N/S and E/W, in the other you separate trains from the rest).

 

And I don't agree that you don't need as many cross streets. Reducing the amount of cross streets in urban areas will ultimately reduce efficiency because you reduce the options to get from A to B, and it will make the average path you have to travel longer. Not a good option, I think.

 

And you're still left with the challenge to connect the N/S elevated roads with the E/W roads at ground level. Ramps at every street corner are the only way, but this needs a huge amount of space, which in many cases is already occupied by buildings.

[npts2020]

CaptainPanic; If there was an efficient nationwide transit system already, I would likely not be arguing for advancing the road technology in as sudden a change as I do. The last time I was in Europe was about 1980 and I can tell you that even then public transit there was much better than it is in most places in America today. There are some local systems that have their act together and provide good service but many do not and the national grid is pathetic for passenger travel. Also the rate of electric train use is far higher in Europe and on local American transit than for freight in the U.S. (I know there are numbers out there but I forget where I saw them). It is true that the cost of rail and interstate highway are comparable (high-speed rail is slightly more though) but I believe that by prefabricating the roadways in sections we can both significantly decrease cost, and increase speed of construction. In urban settings, any construction is going to be both disruptive and expensive so I would think the initial thrust would be to connect different city's transit systems and enable citizens in smaller communities to connect. Space for construction of merges is a problem, too, as you rightly point out. One thing about America is that our streets are mostly wider than European ones, I believe that by running vehicles (the majority anyway) more the size of a Cooper Mini than the size of a Humvee will address much of the problem along with the fact that automated vehicle don't need as much room for safety margins. We will disagree for now about the need for however many cross streets, as I have yet to do any serious consideration of the matter. Here is the link you wanted, it also has a good description of what is and isn't high-speed. Seems like it was the Japanese that did 581 kph not the Germans. The Germans were the previous record holder.

[CaptainPanic]

I agree with the plan of attack: first improve long distance transportation, while you think about cities. Cities are slowly being taken down and rebuilt anyway, so that is the right moment to also improve infrastructure in those urban areas.

 

Interesting link about the speed records! There still exists another wikipedia website that claims that the French are the current record holders. Well... I guess that they're the record holders for something on wheels, while the Japanese actually fly their (maglev) trains