Robocars: The end of Urban Transit


You may have seen in earlier blog posts my discussion of the energy efficiency of U.S. transit. I started that investigation because as I learned how inefficient most transit systems are (due to light loads outside of rush hour,) I realized that ultralight electric cars, enabled by Robocars, are more efficient than any transit system. Who would take transit if a fast, comfortable, efficient vehicle will take you directly from A to B? This drives chapter eight, about:

The end of urban mass transit

(This one gets the people who think they love transit, rather than loving efficient transportation, in a tizzy.)


Trains are circuit-switched. Robocars are packet-switched. Many of the advantages in the networking realm are analogously advantages in the physical world of transport.

That's because while these little electric cars might solve one of the problems with individual urban transit they still leave a host of others. Energy efficiency is simply beside the point.

The things that we care about the most are architectural soundness and sane land-use patterns. Why the heck should I have to use a car or transit at all? Why shouldn't we have neighborhoods where you can just walk to get a loaf of bread or whatever? It's because of parking. Parked cars take up huge amounts of space. Making the slightly smaller helps, but only slightly.

OK, so if you can make these cars fold up to the size of a suitcase, then maybe you're on to something. Otherwise you are totally missing the point.

Robocars solve most parking problems. When not in use, you do want them nearby, but only enough for estimated immediate demand need be near. The rest can be anywhere that makes sense. Not too far (due to efficiency) but you can store them anywhere -- in other people's driveways, triple-parked on roads, or in old style parking lots parked with Valet efficiency. They will move out of the way on demand.

So that makes it easier to have a walkable neighbourhood. Whether people will actually do this is another question. Some will, sadly, walk even less.

Transportation is not something people will do without, even in nice walkable cities, because we live in too many disjoint communities, and won't give them up.

Transit also tries to fix the congestion problem. Even hoards of the most fuel efficient cars/motorcycles can't fix the congestion problem. PRT which is being tried out at Heathrow Airport may at last solve this problem. Small electric vehicles on a dedicated guideway which allow them to skip stops which are unnecessary (no station dwell time) can be fast and move more people faster. Think of it as a horizontal elevator. The issues being evacuation of the cars in the event of a car failure/fire require cat walks on elevated guideways. Safety for all passengers so that they won't be assaulted by the other riders. (panic buttons etc.) Car break downs... let the car behind push the car in front to the next station. And having the empty cars at the right station at the right time. (Think about the end of a ball game and the crowd rushes to the nearest station yet the cars are all idling out at the fringes of the system.)

Robocars are effectively a type of PRT, except they run on ordinary roads (of which there is already a huge supply) and don't have stops or places they can't go. What PRT gets is that at the large expense of building the elevated track, you get some more lane capacity.

Robocars can do a lot for congestion. First of all, once they are ubiquitous, they can pack the roads more tightly, safely, due to the faster reaction times of computers. Thin one and two person robocars can also pack much more tightly than today's cars. Third, robocars can use the entire road network, while human drivers tend to congregate on "main roads." I anticipate a service which is both constantly broadcasting data on present and predicted load/congestion on all roads, and also taking reservations to use the most congested roads in the near future, keeping them from going over their limit.

Just keeping traffic flowing doubles capacity over stop-and-go. Having half-width cars doubles capacity again. Having closer spacing can double capacity yet again.

However, if there is not enough capacity after that it's easy to add buses and trains to the system. In this case, when there is too much demand for a road, a robo-bus is scheduled. All the private robo-cars zoom to a small parking area and the passengers get out and quickly load the robo-bus, which then goes nonstop to a target area. At the target stop, a fleet of single passenger robotaxis awaits, and passengers quickly get into the one with their name on it. Or walk, if they are close.

Brad: I enjoyed these article. Sure make you think about the old CW. Here are a couple of transportation links that you might find useful grist for your mill:

USA Railroad Passenger-Miles per Gallon 1936-1963.

How the Auto, Truck & Airplane Replaced the Train, Horse, and Bicycle.

You might consider this alternate evolution: cybercars are allowed in neighborhood streets at very low speed at first, and they serve to deliver people to high capacity transit stops (such as buses/BRT). So initially they would boost transit ridership a lot. Then eventually they would become dual mode or rail-captive PRT or replace the corridor transit modes.

In that the general problem with "new infrastructure" robocars, that need special lanes or roads or changes to the roads, is that they can thus only operate in a limited area. The transit shuttle you propose provides a good application for a limited area robocar. Other shuttles could also make sense, to malls or stores (which might provide the free robotaxi to shoppers) or to carshare lots, though in the latter case it makes more sense if the carshare car can come to you.

The efficiency of mass transit increases nearly proportionately to the number of passengers. And critically, the number of passengers increases as the size of the system grows, as the number of possible "Point A to Point B" trips increases as the number of stations SQUARED. A small system connects few places, therefore it attracts few passengers and is inefficient. A big system connects MANY places, attracts many riders, and is very efficient.

The best American transit systems (NYC, SF's BART) are equal if not better than a Prius. The world's best system, in Tokyo, uses far less. Japan Railroads East claims about 530 btu/passenger mile according to Wikipedia (actually, they claim in metric as .35Mj/km). Note that Japanese passenger trains are almost entirely electric, and therefore can more easily switch to renewable energy than liquid-fuel dependant cars (even yet-to-be-made-practical plug-ins).

Your argument boils down to "Our transit system is small and weak, and therefore attracts few riders and is inefficient. Therefore, we should keep it small and weak." This is circular reasoning, and you should know better.

Actually, it is my main point that the poor ridership is the cause of the inefficiency. But when you say that "the best american system" is on par with a Prius, I count that as a failure. Because we can make vehicles that are much better than a Prius. Americans, given the choice, chose inefficient big cars, though their tune is changing a bit as gas prices go up. Even if they could not buy those, and had to choose between greener light vehicles (such as small electric cars) and less green transit, they would go for the private vehicles in most cases.

The statistics don't immediately say what we should do. They say what we have attained. And what we have attained is to be on par, or worse than efficient private vehicles.

Note, it is not clear from the JR cite you give if it is "fuel to wheels" or "electricity to wheels." There is a huge difference, a factor of 3, between the two. I do agree Japanese trains are more efficient and better utilized, but I want to make sure that any numbers provided are based on the same method (fuel to wheels or well/mine to wheels) because if you slip in "overhead electric line to wheels" it skews it 3fold.

It is unclear whether JR was taking electrical plant conversions into account. Their sustainability report clearly addresses such issue, but doesn't get into passenger-kilometers. I am not sure where Wikipedia got its number. Either way, this could quickly get into an apples and oranges comparison. One simpler way to look at it is cost. A Prius runs you about $450/month to own and drive ~1000 miles/month over twelve years. Living in Japan, and doing ~1000 miles/month equivalent of train riding will cost you less than half that, around $200/month. Japanese train systems are largely subsidy-free and profitable. American auto driving is not...even a Prius gets to dump its fumes cost-free, gets free parking in many places, and only pays for a fraction of the cost of the roads it is riding upon. Yet it loses on cost 2 to 1. This is a strong indicator to me that the Prius is consuming a lot more resources. Note that the Prius's total cost is only marginally affected by its efficiency. Making it "better" in the future will not lower cost much. Any savings your robo-cars will make by cutting into the $80 fuel fee and the $100 insurance fee will likely be equally offset by increases in the price of the car itself. Robo-cars will happen, but they will not be an energy panacea...just a convenience, especially for the young and old, who cannot drive with our current cars.

Americans love their cars because they are ignorant. It is that simple. I have never met someone who has gone to Japan or any of a number of major European cities and come back with an attitude of anything other than "Why the heck don't we have that?". A car might get you where you want to go a bit faster, but driving pretty much consumes your time while doing it. Using mass transit may take a few more minutes, but since most of your minutes are spent walking, reading, chatting, or just observing the interesting sights, you actually WASTE less time.

A big problem we will have here in the US is that because we not built serious transit systems outside a few select cities, we have sprawled, making it difficult to find high-density places to connect. This is a chicken-and-egg problem, as once you build the transit systems, the areas around stations BECOME high-density.

There is another reason I like transit systems: they are one of the very few things we can build that will last. When I was wandering through the mega-stations at Shinjuku and Umeda, one thing I could be sure of is that those places would be used for generations to come. The same cannot be said of what we have built here.

I have some other thoughts:

1. I think as mass transit becomes obsolete that remaining infrastructure could be used for Robocars. IE in NYC the underground subway lines and stations could become queing zones or express routes if the streets above get jammed up. People might still enter and exit their vehicles at a few stations, and the subway in some way would live on...

2.I think robocars could be subsidized, if that is even necessary due to their cheapness, to serve the poor who don't own transportation in this age. Mass Transit designed for this purpose is usually massively inefficient. Think about those van-bus things circling suburban neighborhoods with less than a dozen riders every day.

I think this will take some time, but will happen. It will be highly political. Look how long it's taken to get the High Line in New York converted to something else.

In quantity, we could build electric tricycles with a fiberglass shell for under $2,000. At first the computer systems will be costly but as they come down in price, I agree that the rides will be cheap enough for the poor. Cheaper than subsidized transit, without a subsidy. Almost too cheap, in that the cost will come more from the taxi company's logistics than the cost of the vehicles and electricity.

Interesting ideas. And Electric bicycles and scooters are doing fine these days.

But you know mass transit can be modernize in the same way. With variable length trains made of electric modules able to recover some energy when braking.

A big advantage of mass transit is the separation of traffic. You should put all the containers on train instead of putting them on lorries.

If you're talking about the Commuter Cars Tango in your charts, it is an inline TWO passenger electric vehicle (ie, one-behind-the-other, similar to motorcycle-style seating) . It is also illegal on the roads in most states (as it has 4 wheels it is considered a passenger car, and threfore requires crash testing to be anything other than a kit-car). That being said, I'd love to have one ;-)

I love your anlysis, of course that could be because I own and drive a 10-year-old Twike (Twin Bike), a three-wheeled Electric Vehicle/bicycle thingy ;-)

In the first year of ownership I put almost 4,000 miles on it, driving it around town, back and forth to work, to drop off Maddy-n-Mia (7 and 2 years old) at school and daycare, tasks that I could not accomplish (and get to work by 8:30-9am and school/daycare pickup by 5:30) using C-U's "Best little transit system"... going on year 2, and while the novelty has worn off, it's hard to have a bad day when people honk, wave and smile at you on the way home ;-)

The Twike is listed as getting 250-600 MPGe, and at the '08 Madison WI Hybridfest 26-mile test loop I calculated 168 MPGe on coal-generated electricity (which soars to a 500+ MPGe on renewable-sourced electricity) -- based on the MPGe formulas at Wikipedia.

Yes, I've sat in the back of a Tango. It's not a great experience, but I should note about that.

Your Twike uses expensive nicads and has a 50 mile range. My vision of the mini robocar uses whatever battery is cheapest (including the cost of energy to move the weight of the batteries.) That's probably lead-acid. As to how much is in a typical car, a subset of cars would be long range (50 miles like your Twike) but many would be only 10-20 miles. The question is the economics of downtime for the vehicle. If the vehicles and batteries are cheap, you give them less range and they spend more downtime charging (though at fast-charge stations.) If the vehicles are rarer and more expensive, you give them more range so they don't have to recharge as much, or more probably you enable them for battery swap.

Lead-acid actually has an advantage. A tray of lead-acid on the very bottom of the vehicle provides stability.

Hmm... awesome about getting to sit in a Tango, bummer that it wasn't really comfy -- Der Twike is a true two-seater that I've slightly modified into a 2.5 -- added a car seat. The Tango probably has a better crash protection tho' (protection from other vehicles). I've altered my driving habits from the full car -- stick to tertiary roads where traffic is a lot saner (same roads I'd bicycle on).

Yep, the video of the Tango carving up that course is very impressive -- shows that a floorboard full of batteries gives you a very low GG... the downside is that it's the weight of a small Camry (2-3,000 pounds if I recall). This becomes pretty wasteful in stop-n-go traffic or stop-sign-every-block style city driving, because as any bicyclist can tell you, most of your energy is spent getting moving from a dead stop... of course going towards inteligent cars that could sense traffic conditions, it's not as big a deal... unless you crash and you've got all that kinetic energy to disperse somewhere.

along the read i noticed the author has presented several disadvantages for the robocar system:
the "american tradition" of car ownership
leaving your things in the car or customizing it with objects
having to wait for a ride which is ideal for your specific trip
potentially having to transfer between different types of vehicles along the way

it seems to me that the solution to most of those exists in the solution he offers for handicaps...

why not simply have your own "sub-car" or "pod" so to speak? one that perhaps could be transfered between vehicles mid-drive (thus always going on the first free vehicle available), one where your put your things in, perhaps even engineered in a way that compacts and follow you around when you don't need it...

I must object to you're statistical analysis. I am here sitting at the 28th addition of the transportation energy data book and it's fun how you skew the data to support your idea. Average rail is about 2577 in that book, but you start off with the least efficient light rail to try and make a point when it's clear that such a system requires passengers (which it has been losing) and hasn't been updated with periodic advancements. In the interest of fair statistics I must insist that your analysis is unsightly. As a point of fact in Europe the national rail system, particularly the new ones popping up in Spain are both highly efficient, and replace airlines. Your concept that somehow urban transit will disappear is also unfounded, both because it's a semi-permanent system and because as seen in the 1970s it's easily more efficient than cars when used.

I talk about the strange average rail number in the latest data book and indeed I am curious about how to reconcile it with the average numbers they give for light and heavy rail. As for your other criticisms, I hope you can be more specific than "unsightly."

My overall conclusion remains the same. The efficiency difference between public transit and new generations of private cars is small, and often negative, and in light of that, we have better hopes for efficient transportation from a fleet of small robots than we do from large shared rail. If you have an argument as to why that is not the case I would like to see it.

That is a great article Brad.
We reached the same conclusion coming from different vantage points.

Have you read the paper by the Earth Institute of Columbia University, ‘Transforming Personal Mobility’, August 2012? (It is co-authored by Larry Burns - a fellow Google consultant to you). In it you will see that for a city with 200,000 vehicles that an optimized autonomous vehicle solution replaces 120,000 of those vehicles in the urban centre with just 18,000 autonomes that are kept busy 70% of the time. The average cost to a user falls from around $16/day owning a vehicle to as low as $3/day - less than a fifth of current transportation costs. Even heavily subsidised there is no way that Transit can compete with those rates - especially when it won't be the public sector that pays for the 'autonomes' (as I call them), but the private sector.

To justify how we could arrive at such cheap rates for the autonomes, here's an implementation scenario that assumes that 'body-out' function is certified safe:
First the taxi and car hire businesses disrupt - they switch to autonomous fleets as quickly as possible. The autonome taxi companies can reduce costs by 14-30% and so human driven taxis will struggle to compete. The car hire companies have no choice but to join-in as they foresee demand for self-drive vehicles will fall very quickly.
Car sharing companies like RelayRides and Zipcar join in the action as they are entrepeneurs and saw this coming years ago (well some have Google as key investors/advisers already). So they add their autonome fleet to the mix. '
But at the same time aspiring entrepenuers like university students, realizing that their cars are typically idle more than 90% of the time, go and buy an autonome or a cheap car suitable for retro-fitting, and hire it out when they are not using it. So they make money by owning a car…… that would catch on.
The cheaper student cars are difficult to compete with the other companies already mentioned as they have larger overheads, and so a balance develops between the autonome fleets where you get a really cheap price from the student model, or a newer/shinier experience from the experienced companies.
Eventually it isn’t cheaper to own a car and rent it out – as the competition has driven prices really low.

What concerns me is that every day government and businesses are making key decisions with significant financial and societal costs that would be much better made witha a cognisance of the impacts of an autonome future. Public transit is a perfect example - with many schemes taking years to construct with revenue pay-back models decades long. This autonome fleet model will likely arrive long before some schemes are even built, and within the pay-back period of most of them. The transit operational and business model will be disrupted by the autonome model and we will probably be left with a number of 'white elephants'. Through my professional work in the transportation profession I am trying to raise awareness to this paradigm shift so that at the very least decision makers ask if due diligence has been carried out against the impacts of an autonome future. That way all parties involved are forced to become more acquainted with this transformative technology - and hopefully at some point will realise the inevitability of the robot revolution.

After all, it might only be another five years after autonomes arrive that we will bave robots capable of driving vehicles and doing manual work either end.


I live in California where it is legal to Split Lanes on a Motorcycle.

I commute 43 miles a day on 405 freeway- Splitting lanes is the only thing that makes my commute worth while.

Will/are robocars be able to deal with this?

Thank you-

This is commonly seen in California, and most teams (for now) have development going on in California, so they all have to deal with it. I suspect most teams would wish it would not happen, since it is more work, but they don't get a choice.

Add new comment