Autonomous Trains

[Chuckanut]

Here's an idea whose time may have come - Autonomous trains. Four times more efficient that trucks, easier to make autonomous than car or trucks, they may be the way more good are moved in the future.

The article is behind a paywall so here are some quotes I found interesting.


https://www.wsj.com/articles/how-to...ogged-infrastructure-robot-trains-11633812291

But researchers who study the matter say that making them fully autonomous could improve their safety and also significantly increase the amount of freight that can be carried on America’s rail network, by making more efficient use of it.

trains—which are four to five times as efficient in energy used per mile as trucks

Most important, trains are on tracks, which means an artificial intelligence train engineer doesn’t have to worry about issues affecting drivers of cars or trucks—no tailgating, lane changes, left turns against traffic, and so on.

But it's not all a bowl of cherries:

Despite the head of steam engineers have built up, automatic train drivers aren’t ready to be rolled out just anywhere, cautions Dr. Dulebenets. Most autonomous trains are built on new and dedicated tracks they do not have to share with other, human-controlled trains. These newer systems tend not to include hazards like highway crossings, where the vast majority of accidents involving trains and motor vehicles happen in the U.S. every year.

 

[CardsFan]

I am confused about the cost savings. Is the elimination of 5 or 6 people (this is a guess by me, if someone knows the true number please chime in) on a freight train significant in the cost of transport? Maybe it is.

It seems to me the true benefit would be from improved tracks and train design, allowing faster transport.

FWIW, could get past the paywall, so did not read the whole article.

 

[OldShooter]

Actually, self driving trucks is a topic of more interest to me but there doesn't seem to be a lot of press chatter about it.

Right now, on-the-road time for trucks is limited by DOT driver rules. It seems very feasible to me that self-driving could double the useful, on-the-road, time of a truck. The impact could be huge. First, inventory $$ in transit would be halved. That is a lot of freed-up capital. Second, the land area where trucks could provide next-day delivery would be quadrupled --- much cheaper delivery than air.

Secondary impacts would not be huge except, of course, for truck drivers. Terminals would still be handling the same amount of freight so no additional floor space or labor should be required. Truck maintenance staff and space would probably be unchanged since the number of miles traversed would be about the same. Possibly demand for new trucks would have an initial dip since the equipment would be used more efficiently. IOW the rolling stock $ inventory would drop just as the inventory $ would drop. But in the end the trucks are driving the same number of miles, so once the rolling stock inventory adjustment happened then demand would return.

Electric trucks are an independent issue, but it is easy to see a future where self-driving trucks pull into "fueling" points and exchange a (huge) plug-in but discharged battery for a full one. Downtime probably in the same range as a current liquid fuel stop.

There is also talk of interconnected truck convoys improving fuel efficiency, whether electric or conventional but that IMO is second-order stuff.

Interesting stuff to speculate.

 

[CaptTom]

They seem to be ripping up tracks and converting them to walking and biking trails everywhere. Once someone builds housing near the old railroad right-of-way you can never get it back.

"Alternative transportation" advocates are always pointing to trains, but I just don't see any way to expand that infrastructure. I think it's a shame, but that's the reality here in the US.

 

[ERD50]

I am confused about the cost savings. Is the elimination of 5 or 6 people (this is a guess by me, if someone knows the true number please chime in) on a freight train significant in the cost of transport? Maybe it is. ....

My thought as well. And it may not be so easy to replace that crew:

(also probably behind a paywall):

https://www.chicagotribune.com/news/ct-xpm-1985-09-27-8503050693-story.html

By United Press International CHICAGO TRIBUNE

September 27, 1985

A presidential mediation board said Thursday that railroad firemen are no longer needed for the safe operation of diesel trains and recommended eliminating by attrition about 5,000 such jobs nationwide.The White House announced the conclusions of the three-member board, appointed by President Reagan on Aug. 30, in the face of a threatened nationwide strike by 67,000 United Transportation Union workers.

''The board has concluded that the time is now, 26 years after the completion of the change from steam to diesel locomotives, to write the final chapter in what was described to be one of the longest, most studied and volatile labor disputes in the history of railroad collective bargaining.''

-ERD50

 

[Sunset]

Eliminating a few drivers per train won't be a big cost saver, but if trains intercommunicated with each other, they could double or more the trains on the rails, while the separate trains acted in unison as needed.

Biggest issue again is the interaction with humans, people still get hit by Chicago Metra trains (commuter trains) quite often where ever this is a car or foot crossing.

 

[GalaxyBoy]

My first thought was that firemen were kept for years after they were obsolete; thanks for the link to the story. Another thing is that cabooses were kept around long after they were no longer needed, as well. The railroad unions were strong in those days; I guess they still are.

 

[Andre1969]

I know it's not quite the same type of train, but I know someone who used to work for the DC Metro (subway). He said that the trains would over-shoot the station every once in awhile, thanks to the computerized controls. And if that happened, they were not able to back up. So if it overshot your station, you'd have to wait til the next stop to get off, and then get on board a train going back in the opposite direction.

As for as fully automated trains, I honestly thought we were pretty much there already...that the trains pretty much did what they needed to, and the engineer was only there to monitor things, in case something went wrong and human intervention was needed.

 

[Midpack]

I am confused about the cost savings. Is the elimination of 5 or 6 people (this is a guess by me, if someone knows the true number please chime in) on a freight train significant in the cost of transport? Maybe it is.

It seems to me the true benefit would be from improved tracks and train design, allowing faster transport.

FWIW, could get past the paywall, so did not read the whole article.

I’d be surprised if this moves the needle much on when it’s optimal re: trucks vs trains vs planes vs ships for freight. The “system” adjusts constantly to optimize when each makes the most sense for cost, volume, lead times, availability, etc. When I was juggling rail vs trucks in my career, rail was always much cheaper per unit but lead times were far inferior, “on time” delivery was laughable compared to trucks, quantities were prohibitively large for many materials, and rail customer service was non existent.

 

[Chuckanut]

Eliminating a few drivers per train won't be a big cost saver, but if trains intercommunicated with each other, they could double or more the trains on the rails, while the separate trains acted in unison as needed.

IIRC, the article I mentioned estimated about a 40% increase in capacity when the trains can communicate with each other.

 

[GalaxyBoy]

... 5 or 6 people (this is a guess by me, if someone knows the true number please chime in) ...

In the US it's typically two, although there is a move to change it to one. Countries other than the US often only require one. US commuter trains typically only use one person in the cab of the locomotive.

 

[skyking1]

Positive Train Control will eliminate most human error accidents. I think a combination of crew plus computerized control is the answer. Airlines have had autopilots for decades, auto landing systems, etc. In a simple track operation it is easy to implement, but has taken far too long to get done.

 

[Andre1969]

Now that I think about it, even the Hooterville Cannonball was down to one crewmember, towards the end. The times, they are a'changin'.

 

[Brat]

Operating a train is not just a matter of pressing go, the weight and characteristics of the cars it is pulling impacts operations. If that 'fine touch' could be automated that would be good.

I agree that the cost-saving won't be in reducing the number of operators but in the efficiency of operation. Trains could travel closer together and switch from one track to another a simpler operation, for example. As it is now each company manages traffic from a central office.

 

[MRG]

IIRC, the article I mentioned estimated about a 40% increase in capacity when the trains can communicate with each other.

+1
I think "the pony" is optimization in the delivery of freight. I'm not sure I understand what it means but many of these systems could/should be rethought instead of old barriers being accepted.