High Speed Rail Falls Behind as Intelligent Cars Begin to Appear

California High Speed Rail

Proposed California High Speed Rail (AP Photo/ California High-Speed Rail Authority)

Boyhood memories of trains and fascination with steam locomotives bind much of the railfan community together and nostalgia brings everyone out for photo shoots when a classic train makes an appearance, but that very same nostalgia should make us cautious when planners propose massive high speed rail projects to reduce travel time between cities and eliminate congestion on our roads. Although those goals are admirable, the rail projects are rooted in the past and ignore the various technologies already appearing in intelligent cars and the rapidly approaching autonomous cars, like those being tested by Google, as potentially far better solutions than trains could ever be. Intelligent and autonomous cars make high speed rail seem outdated before the project is even under way.

Cars are already in showrooms with lane sensors that keep the car from wandering into oncoming traffic or adjacent lanes, adaptive cruise control can slow the car if rapidly approaching the car ahead and even apply the brakes if the driver is unaware of an obstacle blocking its path due to fog or darkness or any other reason. Sensors can also detect overtaking traffic in a driver's blind spot to prevent pulling out at the wrong time. All of this is available now.

Intelligent cars communicating with each other enabling increased highway capacity

Intelligent cars communicating with each other enabling increased highway capacity

Inter-vehicle communication makes cars aware of one another and takes the power of sensors even further because now all vehicles in the area know where other vehicles are and how they are moving. Tests in Europe have shown "road trains" can safely and rapidly move a group of cars driving very close together because the computers can react to changes more quickly than a human behind the wheel. Though not yet available for sale, the technology exists.

These technologies can dramatically increase highway capacity and this is before we even consider totally autonomous cars which Google has shown to be very safe. It was just six or seven years ago that DARPA was running the first competitions to see if any vehicles could complete a course without human driver input, now, Google's cars have driven over 300 thousand miles on public roads and their only accident was when a human took over! The speed of this technological advance is breathtaking. Imagine its capabilities in the next 20 years.

Compare the auto technology to high speed rail, like the project proposed in California. That technology is and will continue to be what is available today when the project gets under way and speeds are no faster than that attained decades ago. The cost is huge, the time span from start to finish is almost two decades and when completed will only serve people in a very specific and limited area. It's hard to know what trains might be capable of in 20 years, but if you begin building now, as soon as it's finished, it's 20 years or more out of date, while auto technology is advancing at mind boggling speed.

Intelligent cars can also continue to eat away at any current advantage the train might have if it were somehow instantly available today. Intelligent cars can speed up the trip to and from the airport and pickup and deliver passengers door to door, not drop them at a single location downtown. The overall trip then becomes faster than any high speed train could manage, and unless you live very close to the station, any potential advantage of high speed rail quickly disappears.

Though politicians love huge projects like high speed rail, enabling them to pass out contracts to favored companies and friends, taxpayers who must ultimately pay, are far less supportive, especially when officials are already having trouble showing it could produce sufficient revenue to cover capital costs, let alone operating expenses.

While most readers here prefer to be in charge of the car they are riding in and not quite ready to hand off driving duties to the computer, the technology is arriving and advancing quickly. It's not hard to see a time in the near future when intelligent cars will get us from place to place far more quickly, and certainly more economically, than any rail project ever could.

Link: The American Interest
Link: IEEE Spectrum
Link: Singularity Hub


  1. Mark L. says

    As someone who has to travel too often, I am looking forward to the day that I can play passenger in my car while it drives me to my destination. I would love to be able to look around at the scenery in the places that I go.

    I say bring it on!

    Mark L.

    • Paul Crowe says

      I agree. Even if you love to drive, there are many occasions where a business or personal trip includes many long and monotonous hours on highways with little visual interest, or if there is, you really can’t look at it without stopping. It also makes those hours productive if you wish them to be. Interesting technology!

    • Luke says

      Even on the most monotonous highways I still enjoy driving more than just sitting there. I must concede however that a computer could do it better and safer than any person could, although it would make me too nervous to watch. I would need to have blacked out windows.

  2. Gidgester says

    Hello Paul,

    I found your article interesting, however I believe that there are a couple of reasons that the coming “road trains” will fall short of delivering the potential safety and efficiency of a well run High Speed Rail (HSR) project. My primary thoughts are outlined below:

    1. True “rail lines” operate with steel wheels on steel rails, hence having very low rolling resistance. “Road trains” will still travel on rubber tires.

    2. The cars in a “rail line” are physically linked together, and the directional control of the entire system is handled by the rails, hence allowing fewer “degrees of freedom” for the system to fail. With a rail line, generally speaking the only externally controlable input is the velocity that the train is traveling at.

    3. The currently proposed “road trains” as seen on SAE’s Automotive Engineering International Online site still have a human pilot for the lead vehicle. As you have mentioned in your blog above, however, autonomous vehicles are coming, so this may be nullified to some degree in the future.

    4. HSR can be electrified -NOW- with currently available commercial technologies. As such, HSR may allow fueling with “greener” alternative energy sources. While electric cars are available, they do have limited range and limited life cycles of their large, heavy battery packs. Building and recycling these cars / battery packs in mass quantities will also be likely to have negative environmetal impacts.

    5. The equipment in a HSR system is designed to last for millions of miles with proper maintenance. Road vehicles have a life-span of roughly 100,000 – 200,000 miles, hence most likely requiring much higher total investment in terms of miles traveled / passenger.

    I would also counter that the government maintained highway system is just as great a sinkhole of public funds as any proposed HSR system could be.

    I envision a future where both technologies may be at play.


  3. GenWaylaid says

    An autonomous taxi system like the “Johnnycabs” in “Total Recall” (the Schwarzenegger one) has the potential to outcompete not just trains but every mode of transport in the 5-500 mile range. Trains, subways, light rail, buses, taxis, and even personal cars couldn’t match the convenience of an autonomous car that can promptly meet you at any place and any time of day and take you (and your luggage) to any destination. It could well be the biggest change in transportation since the personal automobile.

    It may be too soon to count trains out, though. A speedier version of Amtrak’s limited Auto Train would allow riders to cut the middle out of long road trips and avoid car rental.

    Then there’s Elon Musk’s “hyperloop” idea, whatever that may be. Elon says a lot of impossible things, but I’ve learned not to sell him short. If he thinks he can cut a trip from San Francisco to L.A. down to thirty minutes, maybe he can.

    • Paul Crowe says

      The Google cars are a good first step to Johnny cabs and at the rate these cars are advancing, we may see them sooner than we think.

      Amtrack’s Auto Train is interesting, but one problem is the loading and off loading which, according to comments from some who have traveled on one, can add hours on each end. Perhaps someone at Amtrack needs to study queueing theory.

  4. Peter Miles says

    There a big hole in this argument which has been missed. Gidgester talks about using electricity as fuel, which gets close, but the big miss is the fuel cost of powering hundreds of 4000lb vehicles along the highway in a road train vs putting all those bums on seats in a HSR. When you’re on a congested highway take a look around you and look at how many occupants are in each vehicle. I bet the average is around 1.5 to 2. Then take a look at what’s being driven, mostly cars in the 4000 to 5000lb range. That’s a lot of fossil fuel being converted to carbon dioxide etc. Even if they were all electric cars you have to consider how the electricity was generated (coal perhaps?). Unfortunately, nothing compares to the convenience of driving your own car to all those destinations we need and desire to go to, but this is coming at a substantial environmental cost. HSR’s could be a real good way of reducing environmental impact, but they will never be effective until the very real issue of convenient local transportation at both ends of the HSR is addressed.

    • FREEMAN says

      It is much more environmentally friendly to run a fully loaded train than the same number of passengers in autonomous automobiles. Autonomous buses would be cool, though.

  5. MrB says

    Automated auto technology holds promise, but there are several reasons why it does not compare with HSR. 1) ra il is far more energy efficient on a per passenger/permile/per unit of energy basis. If nothing else look at the total drag generated by the number of cars/trucks carrying 2-300+ people vs. one train. 2) few cars are capable of 150+ mph, so you are limited at least by the the lowest common denominator. 3) the difficulty in gaining consistency acros all vehicles Is a huge challenge. Everyday I see cars ranging from 40+ year sold to current on the streets. It will be difficult if not impossible to mandate installation of the technology. And then think of the riots when ver. 2.0 comes out!

    Millions of people daily put up with the same inconvenience when traveling by air by the way.

    All this said, it is a worthwhile technology and should be pursued, but neither is it an overnight implementation, nor can it provide the same efficiency and type of travel as HSR.

  6. ROAD TOAD says