The Train Rider

This post is a continuation in my 'Perspectives' series, which will be written in the first person by a fictional commuter from the future - when autonomous transportation has fully matured. The purpose is to illustrate the ideal future, in which all the opportunities have been seized at the right moments to create the best public transit system possible. My hope in writing these speculative fictions is to demonstrate the possibilities that autonomous transit vehicles will open up for us - and hopefully inspire us to make real steps towards preparing for a similar future.

***

If there is one image we share in our minds about travel in the future, it’s passengers climbing steps up into a train car, ready for a long trip to a far-away city.

No? You don’t think that sounds very futuristic? You think that’s an image from the distant past?

Well, I’m here from the year 2043 to tell you that the future is a surprising place.

My Situation

My name is Yargi. I was named after a character in a popular fiction of my parent’s day, but that’s still before your time so there is really no point in explaining it. I live in the city of Evanston, Wyoming, which in my day has made a name for itself as a host for many medium sized conventions and gatherings – the reasons are too long and varied to put down here. The point is that while the city is only just large enough to be described as a city rather than a town, it isn’t any out-of-the-way cow town like so many other settlements in my state. Of course, the reason the town was founded and the reason it continues to exist at all is the Union Pacific Railroad, which built through the town center in 1869 on its way to Promontory Summit in Utah. It is that same Union Pacific Railroad that continues to run autonomous electric freight trains through our town today, and – believe it or not – it is also that same Union Pacific Railroad that runs a passenger service between Evanston and Salt Lake City.

It actually is a much more comprehensive service than that, with some trains going as far as Cheyenne and Denver, but those trains run only once or twice a day. Trains to Salt Lake City can run as often as five or six times a day, and are very popular among day travelers.

Long Distance Travel and Autonomous Cars

It’s not all tourists and convention attendees that take these trains. I’m neither, and when I go to Salt Lake I prefer to take the train over an autonomous car or bus, and I’m not alone. Autonomous cars can become pretty pricey the farther it is you need to go, due to the lost time the car will suffer as a result of your trip. If I were to take a car to Salt Lake, the car would be earning money (my money) for the 1 hour trip from Evanston to Salt Lake – obviously. But then there is the choice: does the car stay in Salt Lake to wait for me, or does it go straight back to Evanston? If it stays in Salt Lake it could earn some more money for its owners by doing shorter jobs around the city - but that isn’t a certainty, and so some margin of error is built into the price of my fare from Evanston. There is also the issue of the car being an out-of-state vehicle, which means it would have to pay extra taxes for the work it earns in Utah, and other kinds of crazy laws and regulations states force on each other. (These laws are amazingly dense, bureaucratic, and invasive, but no one complains because the average human consumers never have to interact with them – only the robot cars and their parent companies. AI’s are infinitely better at sorting through legal price wars and the governments know this, the end result being an exponential expansion of laws on the books directed specifically at robots. Trains, being owned and operated by a single corporation on privately-owned infrastructure avoid most of this nonsense).

The other option, returning the autonomous car to back to Evanston empty, isn’t ideal either, because then there is another hour of the day, possibly during the prime business hours, when the car isn’t doing any profitable work as it drives back to its home base. And since the car’s owner isn’t going to pay for that lost time out of the goodness of his or her heart, that extra cost gets passed on to me, the traveler.

Fare Economics

(If I wanted to get really into the weeds on the subject, I could bring up commission price and per-mile price fare structuring, which is used by most autonomous taxi companies. This means that each fare you pay can be split into a ‘per trip’ base fare, and a ‘per mile’ fare to be added in on top of that. The ‘per mile’ fare pays for the cost of the ride while the ‘per trip’ base fare is mostly profit for the company. All of this means that autonomous cars are generally more profitable when they are performing many short-haul trips in a day rather than a few long-haul trips; but I think I’ve already made my point…)

Autonomous Buses

Buses are designed to fix this sort of scenario, since they go straight from station to station where there is a guaranteed source of traffic and thus no lost time. The economics of having multiple people traveling in the same vehicle also reduces costs. Combine these two factors and you’ve got an enormous bus industry in 2043 that accounts for nearly one half of all ground-based long-distance travel in the United States. Seriously, the price is so low and the service and amenities are so good (well, at least acceptable) that the only reasons most people have for notusing one is that they 1) are not traveling to a central city 2) they are traveling with family (and having many people in a vehicle reduces the price per person), or 3) they are one of those rich people that owns their own car and can afford to pull it from revenue service for personal reasons.

Or, as a fourth reason, they take a train.

Buses VS Trains in 2043

Since buses are the main competition to train travel in 2043, here is how they compare to each other:

• Trains and buses cost roughly the same - sometimes more, sometimes less depending on the season, time of day, weather, and any other number of things. Prices of transportation have become extremely fluid, and it’s absolutely essential to use a travel-aid app to compare prices before purchasing tickets.
• Trains and buses usually go to the same stations. In the case of my Evanston-to-Salt Lake trip, I depart from the Evanston Transportation Center at the Evanston depot – where local vans and intercity buses congregate in a hub – and arrive at Salt Lake Central Station, which is similarly shared. Not all cities saw fit to place their bus hubs near rail stations which is understandable, since large freeway interchanges often have better connectivity from a strictly road-based perspective – and since many of the rail stations are owned and operated by the private rail companies – but an amazing proportion of cities did, which makes interconnectivity between transportation modes so much easier.
• Trains are not as fast as the bus. They are often close, but rarely – if ever – are they actually faster. Sometimes the difference is significant. In the case of my trip from Evanston to Salt Lake, a bus takes just over an hour to get between the two city centers while the train takes about two hours – nearly twice as long. This isn’t as bad as it sounds because:
• Trains are much more enjoyable. The seats are bigger and have much more leg room. The aisles are wide and tall enough to stand up in. The ride is amazingly smooth and never full of the fits and starts and side-to-side motion that autonomous road-based vehicles submit their passengers to. The windows are big and bright, and the restrooms are amazingly roomy. All of this means that it is much less likely for me and many others to become motion-sick on a train than on a bus, which is important when we spend so much of our travel-time absorbed in our digital entertainments. And speaking of which:
• Trains have better internet connectivity, or at least more consistent connectivity, since it is all handled through the railroads’ own proprietary network instead of the patchwork public and private networks that cover the public – and private – highways.

The Experience

So, here is what it is like to ride a train in 2043.

When I go to Salt Lake I typically go overnight (staying with friends or family or whatever), requiring me to pack at least a medium-sized bag. I keep a small backpack with me as a carry-on, usually for just for my screens and other devises but sometimes for snacks. I like to book my tickets as early as I can, since the prices usually are cheaper the farther out you reserve them. Like airlines in your day, you can gamble that there will be some clearance deals if not all the seats are booked on the day of departure, but then you risk both not having a ticket and/or paying much more for the one you do get. So the general rule is (still) book early.

Now, the way I have phrased it - ‘book a ticket’ - makes it sound like it’s a lot of work. It isn’t. In fact, the only thing I need to do is tell my DA (Digital Assistant) “get me tickets to Salt Lake on May 4^th, in the morning,” or something to that effect. Usually the assistant, who knows my preferences (both by my settings and by my past actions), will know that I want to go by train and will want to depart in the late morning – after 10:00 AM if possible. I’ll still check with her (mine is a her) to make sure she got it right, but rarely do I ever need to change anything. After all, I make this trip several times a year, so it’s not asking my DA to do anything new or different. If you’ve never traveled by train before, your DA may need some help determining what to do; usually it will pull up a list of competing bus, taxi, and train services for you to pick through, with a guess of what it thinks you will like best at the top.

All this is to say that really it isn’t that much harder to buy a ticket in 2043 than it is for you in the 2010’s to ask your ‘Apple Siri’ to answer a question of how tall a building is, or whatever. It is the future, after all.

Integrated Transportation

Perhaps the hardest thing for you of the 2010’s to understand is that transportation is usually included as part of a ticket or booking to a larger event. For example, if I were to buy tickets to a basketball game, my tickets would include some way for me to get to the game, cost included. In the same way that instructions to a sporting event in the 2010’s might say “arrive at door 3, then enter at gate A7, then climb the stairs to section BB and find aisle 4,” the instructions in 2043 usually begin with ‘a taxi will arrive at your location at _____ o’clock”. From there your taxi could take you straight to the arena, or perhaps it will drop you off at a train or bus hub for the last leg of the journey.

In this way event planners are able to control the entire experience you purchase, from doorstep, to event, and back to doorstep. It also means that no event will cause inordinate amounts of traffic. Most cities now employ a very complex traffic-analyzing artificial intelligence suite able to discern causes of traffic congestion. If an event is determined to be the cause of massive delays, it isn’t uncommon for the city to charge the event some sort of fine or retroactive road-use fee. This creates another financial incentive for sporting events, conventions, concerts, schools, shopping centers, etc. to use some basic travel-planning programs to coordinate the arrival of their customers by cars, buses, vans, and trains.

If you’re still not convinced, think of shopping cars. Do you own your own shopping cart? Of course you don’t. That’s something you expect the grocery store to supply for you. Any store that doesn’t do that is just cheap, inconvenient, or totally out-of-touch with their customers. In my day, transportation has become like a shopping cart. If you want to go to an event, you can expect for them to arrange your travel in the same way you could in your day expect for a grocery store to arrange for a shopping cart to be ready for you when you arrive (in my day, grocery stores and shopping cars are almost impossibly rare, but I’m told it is an apt metaphor).

All this is to say that when I book a train ticket in 2043, I can expect for the railroad to send a taxi or van to collect me and my bags on the day of travel. What use would it be to buy train tickets if you also had to waste more time coordinating and buying more transportation from your house to the train station? Why would anyone ever bother with the train if there was all that extra work thrown in?

So anyway – back to my human experience.

Since I like to travel earlier in the morning, the first step in my travel routine is usually my DA waking me up. Having kept track of me for so long, her algorithms know how long I take to get ready for the day – and if I fall behind her projected schedule, she’s always there to speed me up. This is important, because the second step in my travel routine is usually an autonomous taxi or van to the train station, and those do not run on schedules. Instead of relying on a set time and place, the taxi is in constant contact with my DA, giving her updates as to when it will be at my house. This way, when my DA informs me that my taxi has arrived, it isn’t in fact outside the door quite yet – but the moment I set outside, it pulls up to my door like magic.

The Station

In Evanston we still have a historic train station, now owned by the city and used as a transportation hub of trains, taxis, and buses of all kinds. It is a quaint location, full of history and beauty and it makes me proud to be a citizen of my city. It is everything a good train station should be.

It is also a great place to watch trains. Though the platform isn’t very long – about 200 feet long, or just enough to squeeze in 3 train cars – it still provides an excellent view of the yard action. What yard action, you ask? Well, you need to learn how railroading has changed in the last 30 years.

The historic railroad station in Evanston, Wyoming, as it appeared in 2017; in 2043 the fences will have been removed so that it may function once again as a train station

Railroad Operations

Back in the day of steam, the word ‘train’ referred specifically to the cars in the consist. It was just assumed that the locomotives would be switched out so often that there was no point in including them in that term. A Union Pacific train leaving Omaha would run behind something like a 2-10-2, which would have been worked well on the flat, straight prairies of Nebraska. In Cheyenne there would have been an engine switch to something more powerful, such as a Challenger or Big Boy, which could pull the train over the continental divide and navigate around the tighter corners through the mountains. And in Utah there would have been another switch as either the SP or WP would have taken over. Never mind the addition of helper locomotives, coaling and watering stops, crew changes, and so many other things that could have meant an engine swap. Basically, labor was cheap and energy was expensive, so the railroads went to a lot of work to make their operations efficient – using high energy locomotives where they were needed and removing them when they were not needed.

On the old Union Pacific line in the days of steam, a 2-10-2 (Above) would be enough to pull a train across the flat plains of Nebraska, but a larger locomotives like the 4-8-8-4 (Below) would be required for the same train traveling through the mountains of Utah and Wyoming

In the era of diesels, this became less of an issue. Diesels didn’t have all the energy and maintenance costs of steam locomotives to the point that energy became cheaper than labor. Cutting locomotives into and out of trains became more expensive than letting a train run over the whole length of the line with an excessive number of locomotives. Soon it became clear that running many smaller trains also meant more labor (more man-hours) than running fewer longer trains – and so average train lengths swelled while the number of trains per day remained stagnant or in some places was even reduced.

Rail fans went from being  able to watch a wide variety of train types running fast and often to being treated to watching generic consists of infrequent but massively long unit trains. It must have been hard to be a rail fan in the run up to automation.

Existential Crisis

Then, one year, autonomous vehicles became a thing. Cars started driving themselves and – more importantly to the railroads – so did trucks. To call this a devastating development would be an understatement. It was a complete and total gutting of all of railroading’s core business. With the labor costs of an autonomous truck practically zero, and with operating hours rules effectively abolished (no mandatory rest-time on long haul trips), autonomous trucks were able to undercut rail rates so badly that many outside observers wondered if the rail companies could ever become relevant in the transportation ecosystem again.

Obviously trains had to become autonomous as well. A crew of two highly-compensated human beings cannot compete economically with a fleet of robots, no matter how many tons of freight are on their train. Longer trains usually meant slower trains, and that was no longer an operating advantage; previously, slower also meant cheaper rates, but since autonomous trucks could undercut that cheaper rate with an express delivery service, the added time of transport by rail became an embarrassment.

So, scrambling for their very existence, the big railroad companies threw themselves at converting their operations from old-timey-railroaders to robotics – and the results were the biggest revolution that industry had ever seen in all its 200-years. A few things remained the same: trains still ran on tracks, the rails were still a bizarre 4-feet 8 ½ -inches apart, and a train still consisted of many cars coupled together. But almost everything else had to change.

I’ll spare you the whole story of technological developments with all its false-starts and dead-ends and jump to my present day.

Autonomous Freight Trains

An autonomous freight train is amazingly short by 2010’s standards. Rarely do trains run longer than 30 or 40 cars, and even more rarely are they anything other than standardized shipping containers. Each of the freight cars is a robot in its own right in that it knows what it’s cargo is, knows the G-force loading its cargo can tolerate, knows where it is and where it is going, and can order the locomotives to slow down if it predicts that the G-force loading will become too much. It has its own self-contained braking system in case of emergencies or unusual circumstances, and has its own wireless communication systems to speak with the other cars and locomotives around it and to transmit messages from other cars or locomotives up and down the consist.

Locomotives can be found anywhere in the consist, but are usually at both the front and back ends. The locomotives resemble big square bricks placed on a shortened diesel-electric locomotive frame. Inside the big square brick part are hundreds of thousands of battery cells, providing both the electric power to propel the train forward as well as the massive dead weight required to gain traction on the smooth steel rails. Batteries and locomotives were such a perfect combination that the moment battery costs dropped enough to make them financially practical there was a tidal rush to switch the entire fleet. Other power sources – diesel, natural gas, hydrogen, etc. – never stood a chance. The extremely low maintenance of batteries, the recapturing of energy through regenerative braking, the amazing amount of horsepower, and the hyper-efficiencies in energy usage made the other power types just as noncompetitive as human-operated trains were in the face of autonomous trucks.

The ratio of locomotives per car changes based on the type of terrain the train is running through, in just the same way that steam power used to be switched and optimized with the terrain. Rarely does it fall much below 1 locomotive per 10 cars, since all trains must accelerate and decelerate no matter how flat and straight the ground. On a train of 30 cars in Evanston (in the mountains), it isn’t uncommon to see three locomotives in the front and 2 in the rear, for a ratio of 1 locomotive per 6 cars. When operating in this way, the locomotives remain in constant communication via wireless relays up and down the consist so that they run in the absolute most efficient manner possible, in the same way that a dual-motor Tesla vehicle would in the mid 2010’s. The arrangement of locomotives is also highly dependent on the route; for a train headed mostly downhill, it is customary to see most of the locomotives in the back of the train, serving as the brakes for the entire consist while at the same time charging their batteries via dynamic braking. In this way they will be charged by the time they reach the bottom of the hill and will be ready to pull another train up the same hill. Friction brakes are still installed on all the cars, but these are rarely used. And since both the brakes and the communications systems are all independent and self-contained, there are no connections between cars other than the mechanical steel couplers. This means that locomotives can very easily cut into or out of a consist whenever they need to.

Shorter trains are not the result of lower amounts of freight moved by rail, but are the way that trains are able to move more today (my day) than ever before. As late as the early 2020’s, a freight train from one city to the other would have as few as one departure per day, but would use a really long consist to handle all the traffic the railroad had captured between the two cities. If your freight arrived at the yard an hour after that one train left, your freight would have to sit uselessly in that yard for another 23 hours. With shorter trains the frequency can be increased dramatically. Now multiple trains per day serve city pairs that were previously once-per-day (or less) operations.

More departures per day means more trains on the track. The national rail network has become swamped with the sheer number of separate trains running at impossibly close intervals. Every siding gets used multiple times per day and many lines that would not ever have been candidates for double-tracking before are being upgraded to allow bi-directional operations. Gone are the days of crew being the economic limiting factor - now time is, and railroads are striving to wring every last operational efficiency from their physical assets. Tracks should not sit empty for hours on end – that’s a waste of money! Send more trains down that line! There should be trains flowing down those tracks like there are trucks flowing down the lanes of a highway – truck after truck after truck.

In some sense, I guess it can be harder to be a railfan in my time than it was even in the 2010’s. The human element is completely missing from railroads, and it isn’t always evident what the robot trains are transporting and why they are doing what they are doing at all. But to someone who has studied logistics, swarm intelligence, and systematic traffic flows, watching the autonomous electric locomotives shepherd their flocks of cars over the rail lines from one junction to the next is like watching a pure expression of efficient transportation in the same way that astronomy is like watching a pure expression of mathematics and physics. Without the corrupting influence of a human hand or mind, and in the confines of the closed-system of railroad operations, ultra-efficient transportation plans are formulated, evaluated, and executed in the eerie electronic silence of ultra-intelligent robots.

All of this is to say that when I stand on the platform of the Evanston train station, I see a lot more than a bunch of graffiti-covered freight cars baking in the sun. On the tracks farthest from me, big boxy locomotives without cabs, walkways, or even crew ladders charge from long, robotic charging cables protruding from enormous battery-banks. The batteries buffer the charge from the grid so that the locomotives can take an enormous amount of charge very quickly without causing blackouts throughout town. This fast charging enables the locomotives to spend less time sitting idle and more time working, which is good since each locomotive costs well over $10 million (some of that is due to inflation, the rest do to the pay-forward nature of a battery-powered economy).

Every few minutes this silence of charging locomotives is slightly interrupted by the almost equally silent approach of a train, marked only by the hum of cooling fans beside the traction motors and the occasional screeching of steel wheels grinding around sharp curves on the steel track. Trains do not wait outside of the yard for a yardmaster’s clearance; instead, the yard computer will have cleared a staging track for the train while it was still many miles out of town, and the train will barrel in at full-speed from the mainline. Once stopped on the staging track, the old locomotives will dump the rest of their free charge into the batteries in the railcars, which will apply their own parking brakes while the locomotives are swapped out. As the old locomotives are uncoupling themselves from the train, new freshly-charged locomotives will already have positioned themselves to swoop in and couple onto both ends of the train with as little delay as possible. Evanston, being near the top of the Wahsatch grade, is one of many locations along the mainline where trains change the number of locomotives in their consists. Trains headed into Evanston from the east arrive with more locomotives than they leave with since more power was required to climb the steep hill from Ogden. These extra locomotives ‘deadhead’ back to Ogden on westbound trains, where they use their regenerative braking capability to charge themselves and slow the westbound trains down the long hill; by the time the arrive in Ogden, these extra locomotives’ batteries will already be half-charged, cutting the amount of time it takes for them to prepare for another eastbound trip.

Once the locomotives are attached to their train, there are a few safety checks the entire robotic consists needs to perform before leaving, which usually takes just a few minutes. Wireless communication between the multiple locomotives needs to be established over two separate channels  - one extra in case of emergencies – while a third and fourth channel is used to establish more communal links between the locomotives and the railcars. All communications are handled wirelessly, after it was shown to be just as reliably effective as physical wire connections but without all the extra equipment required to establish such connections between cars and locomotives automatically. Railcars with certain speed restrictions will verify that the locomotives are aware of the restrictions, and individual brake tests of each car will occur while the stronger locomotives’ brakes hold the train in place. Then, once everything is in order, the train will slip silently out of its siding and back onto the mainline, bound either for Ogden or Green River before needing another swap. In total, the process is usually complete in under 10 minutes.

The Economics of Passenger Trains

So, in this landscape of fast, frequent, and highly efficient freight operations, how do passenger trains fit in? Very simply, really. The logic goes like this: (1) Fright operations have become so flexible that it adding in extra passenger trains doesn’t cause significant disruptions to either the freight or passenger trains; (2) Freight operations already pay for the maintenance and upkeep of the tracks; (3) when train tracks are empty, they are not earning any revenue; (4) Passenger trains are lighter than freight trains and do not cause significant wear onto the tracks; so therefore (5) any money earned by passenger trains is just extra revenue, using an asset that otherwise would not be earning revenue.

To expound a bit, rail companies realized that, so long as a passenger train doesn’t cost them money, it has the potential to make them money. Freight trains are still the priority, and always will be – but on lines that have the extra capacity (due to double or triple-tracking projects), there is the potential to earn slightly more revenue by filling in that excess capacity with something. In the same way that airlines and cruise ship companies of the 2010’s would sell last-minute tickets for extremely low rates (so as to gain at least partial revenue on an unused asset), passenger trains are the way the railroad companies try to minimize their losses on an established investment. The tracks will cost the same to maintain regardless of if a light-weight passenger train uses them or not, so the surest way to lose money is to not run any trains. If you do run trains, you either make money or you lose money, so the probability of losing money is less than not running any. Then, once you build up an established customer-base and develop an economic ecosystem around your transportation service, the probability of losing money continues to shrink, since there is a higher and higher likelihood of your service being in demand.

Being a stop-gap economic tool does not always have its advantages for the passengers, however. For example, when a railroad’s traffic grows faster than its capacity, the first thing to cut is the passenger service. On my Evanston-Salt Lake City route, for instance, there is usually a two or three-week-long hiatus of passenger rail service in the early fall, as harvests from farms and orchards overwhelm the railroad’s capacity. Those shipments are much more lucrative to the railroad than passenger trains, and so any reservations you may have made for train tickets on those days are automatically switched to a bus or cab company (with a notification, of course). It can be slightly inconvenient and disappointing, but it isn’t ever deal-breaking. I don’t ride the train out of necessity, after all – I ride it out of convenience and preference, so I’ll tolerate the slight unpredictability once a year.

The Train Cars

Like the freight trains of my day, the passenger trains are both battery-electric and autonomous. Also like the freight locomotives of my day, they are not much to look at – pure function without a second thought given to form. There is some sleekness to them, but that is due to aerodynamic considerations, not styling. The best description I could give in a short paragraph is probably referring you to the IC3 trainsets from Denmark, which are famous for their cyclops-like appearance:

Amtrak testing an IC3 Trainset at Fullerton California in 1996 showing its cyclopes-like front (Above) and its streamlined exterior in revenue service in Israel (Below) 

Like the IC3, each autonomous passenger railcar uses an exoskeleton structural scheme, leaving the inside of the train open for passenger spaces. Also like the IC3, this scheme applies to the ends of the car, creating the look of a cyclops. This design is very handy for when multiple cars are coupled together, as the end faces of each car is able to open inward, creating a seamless passage between cars that even wheelchair-bound passengers can easily navigate. But, as I said before, the trade-off is a mostly unattractive exterior that makes the old-timer railfans groan and wish for the old days.

The interior of two IC3's when coupled together

Unlike the IC3, the area below the floorboards is a huge battery bank rather than an internal combustion engine with fuel reserves. Also unlike the IC3, each railcar is only a single railcar, not a set of three. Lastly, because the battery bank is below the roll axis of the car, each autonomous passenger railcar is designed to passively tilt into the curves, using the weight of the batteries to swing naturally outward on curves so as to limit the lateral forces on the passengers inside.

Since each passenger car is designed to be completely self-sufficient, each car has roughly the same design. There are no ‘dining cars’ or ‘observation cars’ or ‘coach cars’ – each railcar is a hybrid of all types of design. Large windows that curve over into the ceiling are standard, like an observation car, while seating is in standard rows, 2+2, like a coach car. These benches can spin on a central axis so that they usually face forward, but in a pinch they could be made to face each other, like in a dining car (the standard tray table is not attached to the back of the seat ahead, but instead is attached to the armrest of your own seat, like a folding desk in a college lecture hall; if all four seats deploy their tray tables while facing one another, they are designed to lock against each other and form the impression of one large dining table). Every railcar has a restroom, a shelf for luggage and/or bicycles, a space for wheelchairs, and a (very) small kitchen area with a type of microwave oven for meals. On most trips there is a crew of one person, called the Rail Host, who checks tickets, directs passengers, serves meals that are standard, sells and serves meals and snacks that are paid extras, and generally does everything else that needs to be done while the train is moving. If you’re friendly with them, they’ll allow you to use the small kitchen area to warm up your own snacks or meals if you didn’t buy a travel package that included that stuff. When you’re a cheapskate like me, it pays to be friendly.

Turning seats on a Chinese High Speed Rail train so that all the seats always face forwards

Boarding the Train

In the United States of America, we used to have a pretty mixed-up system for platform heights at train stations. In the northeast, platforms are built to the height of a train car’s floor or 51 inches above the top-of-rail, allowing for seamless level boarding. Everywhere else level boarding wasn’t considered all that important and platforms were built at ground level, or the same level as the top-of-rail. This meant that passengers had to first climb up onto a step-stool placed by a conductor on the ground just to be able to get to the first step of the train car, which were usually about 18 inches in the air.

Level boarding is standard on the North East Corridor between Washington DC and New York City (Above); stations beyond follow a variety of standards, such as a platform at track level, in Alexandria (Below)

More recently, a platform height of eight inches became popular, which allowed passengers to forego the step-stool and step directly up to the car’s first step. The type of train car matters significantly too, as some commuter rail cars have doors that are 25 inches above the top-of-rail, and these require a special ‘high block’ section of a platform to be constructed at this height to allow wheelchairs to be wheeled into the car via a bridge plate rather than being lifted into the train by an external lift.

 A high-block section of the platform (Above) can be used to turn a low-level platform into level-boarding at one door, using bridge plates; a wheel-chair lift (Below) can be deployed from certain railcars, but this equipment takes time to set up.

Perhaps no place was better at illustrating these problems of platform heights than Salt Lake Central Station, where two platforms of odd heights caused some severe logistical issues. First were the platforms designed for the FrontRunner commuter trains, which had a standard platform height of 8 inches on the ‘low block’ portion and a unique platform height of 25 inches for the rest. This was done to allow level boarding in the unique commuter rail cars used by UTA while also allowing for regular cars at the front end of the train. The other platform was an 18-inch platform used by Amtrak Superliner cars, which are also bi-level cars that board on the lower level. When considering that Superliner cars comprised the only vehicle in regular intercity rail service to Salt Lake City, it may have seemed like a good choice to make the platform to be level with this particular height – but it had its tradeoffs. Amtrak trains would often pull private cars behind them which would have floor heights of the standard 51 inches; these fortunately often had stairs which would match up closely enough with the 18-inch-tall platform, but not without leaving a dangerous gap. Furthermore, some historic and freight equipment were too wide to pass by a platform any higher up than 8 inches, meaning that these types of trains were banned from using the platform tracks altogether. A famous example is when the Union Pacific historic steam locomotives came to Utah they would more often stop at Ogden and run straight through Salt Lake City, even though Ogden is a much smaller city. Ogden had facilities that could accommodate the steam engine (a 0-inch-tall platform) while Salt Lake City’s platforms were too specialized for Amtrak-only operations.

In 2043, we have this all figured out, and it really is very simple. All platforms are either zero or 8-inch-tall platforms, and none of them have a high-block section. What we do have that you didn’t have are robotic stairs that autonomously pull up to each doorway like a gangway to an airplane. These stairs are wider than the railcar’s on-board stairs ever were, and are never as steep as those had to be – meaning that even though there are still stairs up into the railcar it is much more comfortable and safe, and very nearly approximates the efficiencies of level-boarding. The stair-bots are also much more flexible than a set of dumb stairs on wheels because they are able to adjust themselves to the subtle variations in height that occur when the track settles at a different rate than the platform. Each step can be adjusted up and down along a track, allowing for the stairs to serve any height of floor they may be presented with, be it the standard 18, 25, or 51 inch heights or even something weird like a 23 and ½ inch, or whatever – there will always be a perfect gap-less transition between the platform outside and the railcar’s floor inside. When a less-able passenger is unable to use stairs, they are directed to use the small lifting platform that deploys out the side of the robot stairs which is usually used for moving luggage and restocking the train. When the stair-bots are not in use, they roll themselves away from the train cars and find some place out of the way to charge or power-down, leaving the platform bare and completely accessible to train-watchers like me.

A mobile staircase used for boarding aircraft; autonomous stair-bots are similar in appearance, though not as tall.

Now, you may think that autonomous robots are much more expensive than building a slab of concrete to a specific height – and in the 2010’s you’d be right. But in 2043 robots are a dime-a-dozen, and very permanent things like concrete slabs cause much more concern for us. After all, once you’ve built a ‘high block’ platform any taller than 8 inches high, that track becomes unusable for wider-load trains, such as the container-carrier cars that have become ubiquitous on nearly every autonomous freight train. Why go through the expense of building something that has a trade-off when there is a solution (flat platforms with stair-bots) that doesn’t have any? Flexibility and maximum-use are the rallying cries of commerce in 2043, and so our infrastructure has become a reflection of that.

Riding the train.

Hardly any time passes between my boarding and the train’s departure. My ride to the station – a well-connected autonomous taxi – usually has dropped me off with only a few minutes to spare. I’m told that since I reserve me tickets in advance, and since the autonomous taxi reports to the railroad when it has picked me up from my house, that the railcar would wait for me should something go wrong and the taxi deliver me to the station a few minutes late – but so far I haven’t had to test this scenario. The car accelerates quickly – the way that any good battery-electric vehicle should – but not so fast as to cause chaos inside the car. At this point passengers are still standing and arranging their bags on the shelves or under-seat racks, and it is rare for me to see anyone lose balance because of the train accelerating too quickly. The side-to-side motion is also highly mitigated by the enormous inertia of the battery pack below our feet, but it is always wise to hold on to something as you walk around the railcar.

Like the Danish IC3, a large window graces the ends of the railcar, and railfans often spend an entire trip standing in a group behind those large reinforced panes of glass. The various sensors that allow the car to travel autonomously – the radar, the cameras, the ultrasonic detectors, perhaps a Lidar unit or two – are spaced around the large window, but are still set into the large front door; this allows them to be serviced or cleaned or replaced easily at stations, as the doors fold into the train when opened.

Very often the freight train ahead of us is visible through the front window, and sometimes the passenger railcar will pull up extremely close behind it. In the old days of air brakes spread out over multiple cars, stopping a train was a very imprecise science, and as a result trains were never allowed to follow one another within sight distance of each other. Instead they relied on wayside signals to indicate if the line ahead was clear or occupied as part of a fixed-block system. Often these blocks could be many miles long, and trains would be forced to be spaced many miles away from each other, resulting in an inefficient utilization of track-space. In 2043, tracks are considered an extremely fundamental recourse; if a train isn’t operating over those tracks every few minutes, then what reason is there to be operating a railroad instead of a regular asphalt road? Moving block systems, such as Positive Train Control, provided an improvement on the fixed-block system, but fully-autonomous trains improved equally upon the moving block systems with spectacular results.

An illustration of the types of communication networks required by an initial PTC system in 2018

Electronic braking and instantaneous communication are key. Every train along the line remains in constant contact with the trains both ahead and behind it by means of the old radio towers and frequencies used in the original PTC systems. Each train will know its precise location and speed and send that information out to the trains ahead and behind no matter if the trains are 100 miles or 100 feet away from each other. If a train does need to go into emergency braking, it will be relayed to the train behind right away. If it can help it, the train ahead will collaborate with the train behind to brake at a rate slow enough so that the train behind can avoid impacting the braking train. If not, it falls on the following train to brake as much as possible so as to mitigate the impending collusion – but those situations are extremely rare in the real world. Braking can be achieved without any delay and at very precise rates due to the electronic communication between locomotives and cars, as well as the independent electronic brakes (and batteries) aboard each freight car. Gone are the days of an engineer dumping the air pressure in a lead locomotive and waiting up to half a minute before the reactant pressure drop reached the end of the train; in 2043 we keep our latency low in every application, just out of habit.

What this means is that, more often than not, our passenger railcars keep themselves within sight distance of the proceeding freight train. Passenger railcars have better braking capabilities, so unless the freight train ahead of us hits a brick wall and stops instantaneously, the railcars will be able to stop long before the freight train will. (Landslides along the tracks could act as a brick wall of sorts, and it is for this reason that autonomous high-rail trucks often drive ahead of any train on a line that has been quite for a few hours, just to be sure nothing is blocking the path.) A close following distance is also beneficial from a safety point of view, since the trains will not only be communicating with each other via radio waves, but the following train will be able to verify these communications with its vision and radar systems.

Combining Railcars

A sad event in the trip of a railfan is when the car you are riding in comes up behind another railcar – usually stopped at a station, but also potentially while in motion – and couples onto it, forming a longer train. This happens when a train trip begins in a sparsely-populated area and ends in a densely-populated area, and more passengers join the train as it moves along its route. Why send a full train all the way out where it will be mostly empty? Why not space out the railcars along the route so that the train picks up cars at the same rate it picks up passengers? In the old days, cutting-in cars on a passenger train meant many long minutes (or potentially hours) of delays as the locomotives detached from the consist, found the new car, coupled it back in, connected up the brake and utility connections, ran checks, and so on and so on. In 2043, the cars are autonomous and connect to each other physically with only a dumb automatic coupler meant to distribute kinetic inconsistencies throughout the entire consist. This makes the process as quick and easy as simply pulling up behind a new car and latching on – and, as said before, this can even be done while the cars are in motion, meaning exactly zero time is lost by adding or subtracting cars. When a new car is added, the Rail Host (car attendant) is required to be present as the doors between the car fold open, just in case any passengers fail to get out of the doors’ way in time. It takes all of twenty seconds for the doors to fold open and lock themselves into their new positions before passengers are allowed to travel between cars – and once that is done, the train has a new front end with a new crowd of railfans treated to the best view, leaving the old ones to either join the back of the new crowd or return to their seats.

The Journey

It isn’t as though the seats are bad, though. As mentioned before, each seat is larger and more comfortable than anything that could be found in a car or bus or any road-based vehicle, and it comes with all the amenities the 2043 traveler has come to expect. Some people prefer train travel for the smoothness of the ride and sink right away into their immersive games and artificial realities that have become increasingly not-so-artificial. But many people still travel by train for the romantic aspects of rail travel – of seeing the countryside from a pristine and unencumbered point-of-view, of traveling through a landscape as a part of it, rather than blasting through it like highways and freeways so often do. Railroads still travel through the centers of the cities they pass through, and with each town you pass you get a sense of what life is like for the people who live there. You see people of all types – the rural and the urban, the rich and the poor, the old and the new. You see that these people are not your political enemies and are not a danger to your society, but that they are a part of what makes the country we share so great and so strong.

It may seem strange to many people of your day to consider trains as the future of inter-city transportation. Even in my day when we fantasize about the future we see it as full of new technologies that disrupt our entire society to its very core. Sometimes these speculations are correct – artificial realities and immersive computer simulations are changing the way we interact with our realities. But much more often these grand predictions of new disruptions overlook the human aspect of a futuristic and modern society. Humans are social creatures and must interact with each other to find any meaning in their lives. We must work together, play together, and travel together. We seek connections not just among the present, but also with the past and the future – we want to feel like we are the continuation of a great work by our ancestors, and that we are carrying it on to a new usefulness in the next generation.

Trains provide us with these connections in a way that other forms of transportation cannot. Trains are larger than buses and can carry more people – building connections to the people around us. Trains move through the landscape rather than hopping over it as aircraft do, building connections to the real world around us. Trains travel routes steeped in history, following the old settlers’ trails and passing through historic landmarks and towns at a rate roads cannot match, building connections to the past. And trains continue to innovate and employ new technologies that will continue step by step along the long road of progress, connecting us to the future. Not every journey should be made by train, but I recommend to everyone I know to get aboard a train at least once in their lives, to feel those connections and to share them with the people they meet in their travels. I’d hate to conclude this essay by saying something cheesy like ‘life is a journey, so travel by train,’ so I won’t. Instead, let me conclude by simply saying that good ideas never die, even if they go out of fashion for a while. Trains helped build the world we know around us, so as long as we wish to live as we do, trains will always have a use. So, if they’re here to stay, why not make them good? Why not make them great? In my 2043, we have, and we are all better off for it. What happens in your 2043 will depend on you, and how hard you work at creating a world that is worth living in.

Best of luck to you, and I hope I find you out riding on the high iron!

Share this post