Before I got a license and a car, getting to and from high school was an ordeal. The hour-long bus ride was awful, as one would expect when sixty adolescents are crammed together with minimal supervision. Avoiding the realities going on around me was a constant chore, aided by frequent mental excursions. One such wandering led me to the conclusion that we high schoolers were nothing but cargo on a delivery truck designed for people. That was a cheery fact to face at the beginning of a school day.
What’s true for a bus full of students is equally true for every city bus, trolley, subway, or long-haul motorcoach you see. People can be freight just as much as pallets of groceries in a semi or a bunch of smiling boxes and envelopes in a brown panel truck. And the same economic factors that we’ve been insisting will make it far more likely that autonomous vehicles will penetrate the freight delivery market before we see self-driving passenger vehicles are at work with people moving. This time on Automate the Freight: what happens when the freight is people?
Running Scared
There’s a saying in the investment game: the trend is your friend. Investors have a lot of sayings like that, mostly aimed at convincing themselves that they have the slightest clue about what’s going to happen with a given company’s stock, when it’s all pretty much up to chance in the absence of inside information. The trends in investing patterns can still be useful, though, because they tend to point out how people are thinking. That can be a handy tool for investors and non-investors alike.
So it was with interest that I stumbled upon an article about the potential impact of self-driving cars on the airline industry. On the face of it, you’d think that there would be no risk to airlines from cars, especially in the United States. But the article points out that US domestic carriers like Southwest depend on routes that average 757 miles (1218 km) and last about two hours. People gladly pay for these flights, enduring the indignities of modern air travel that begin hours before the flight and potentially end days after landing in the form of jet lag. And they do so precisely because the alternative is worse – to most people, at least; I for one vastly prefer long-distance driving to flying. I’m weird like that.
But fast forward to a future where fully autonomous vehicles are an everyday thing, and battery capacities have increased enough that an electric vehicle can cover the same distance as the average Southwest flight. We’re actually not far from that now – a Tesla once did 670 miles on one charge. Or, forget the batteries and consider an internal-combustion self-driver. Either way, a vehicle plying the roads is obviously not going to beat a jet airliner for speed, but it doesn’t have to because it can drive all through the night. And the passengers, unencumbered by the need to drive, can simply sleep the trip away, to arrive in the morning as if by magic, fresh and relatively unharried, at least compared to their TSA-probed counterparts at the airport.
Not convinced that airline executives should be quaking in their wingtips at a self-driving future? Then take a look at this map:
Each circle is about how far one can drive from San Francisco, Austin, or New York overnight – roughly 11-12 hours. Yes, it’s difficult to drive in the ocean, and these are straight-line distances that don’t take the location of highways into account. But still, it shows just how much of the country can be covered with a reasonable overnight drive. And the promise of having slept through it all without having been sealed in an aluminum tube full of unhappy people and their germs might just be a reason for panic in the corporate suites of regional air carriers.
The Night Bus
The idea of sleeping the night away while traveling might sound like it would need to wait for true autonomous vehicles, but it’s actually a service that’s already available. An overnight bus service called Cabin was launched in 2017, offering a premium overnight travel experience that whisked passengers between San Francisco and Los Angeles while they slept. The idea was to use traditional human-piloted motorcoaches outfitted with sleeping compartments falling somewhere between cozy and claustrophobic, depending on your leanings. You’d get on the bus late in the evening, zonk out in your rack, and hop off the bus as fresh as a daisy in the morning.
As anyone who has experienced the marginally maintained roads of nearly every major city in the US can imagine, it didn’t quite pan out. The experience of jolting through the night to the roar of a mighty diesel was less than restful, and Cabin was forced to shut down operations to retool its fleet of sleep coaches. But the fact that they didn’t just fold completely shows there’s still demand for a lower-cost regional transportation option, and that offering an overnight trip is attractive to enough people to make the idea worth investing in.
People Moving on Demand
As for a future, fully autonomous version of Cabin, I’d say that self-driving motorcoaches carrying large numbers of passengers on regional routes are not likely to catch on, at least in the USA, for the same reason that bus lines and passenger rail have never caught on in here: lack of flexibility. Being stuck showing up at a bus depot at a specific time to take the sleepy bus isn’t much different than having to show up at the airport two hours before a 90-minute flight. If a plane offers a choice of departure and arrival times, chances are good that people will choose to fly rather than sleep-drive.
But, imagine smaller self-driving vehicles, perhaps the size of a large van, with four or so bunks and a small comfortable seating area that could be booked for travel on demand. Businesses might jump at the chance to have teams taken safely to meetings overnight, or to use travel time for working. Families might love to have an autonomous van come up the driveway for a pajama-clad trip to Disneyworld. College students could team up for a cheap spring break trip, pool long rides home for the holidays, or just go on a road trip with consequence-free drinking and driving.
All that said, from the point of view of the passengers, where the vehicle is driven by a computer or by a human matters not. All they know is that they’re not driving, and they’re free to nap or read or work while the miles peel away. But to the company providing the service, eliminating the costs associated with hiring large numbers of human drivers is probably a good argument for investing in autonomous vehicles. Providing the freedom of sleeping, working, or partying during the trip would just be a way to add value and attract the customers needed to justify the expense of automating the people-freight.
Anyone else remember there is this thing called Amtrak? Which loses money.
Even for regular freight, trains are far more efficient, never stop except at the stations, and can have large spaces.
There is also the question if Self Driving Cars can work in the rather different rural areas that aren’t mapped to the centimeter and have different condition.
The one key here is with a car with self driving mode, you will have your car at the destination with your baggage. No checking it or having to rent a car or get a taxi.
The article is talking about a minibus or a van that is upholstered for sleeping, not for driving around town during the day. It has the same issue as camper vans – plus the fact that this is supposed to be a night-driver between cities and you can’t hog it for personal use because it has to charge up during the day and then run to other customers.
It still drops you off somewhere – just not at the bus/train station. It takes you from door to door, which “solves” the last mile problem by the letter of the law rather than the spirit of it.
Of course you can sleep in a Model S that’s driving itself through the night, but that’s not very comfortable, and that goes back to square one.
Yea, its amazing how there was no mentioning of sleeper train cars in this article. Perhaps the author hasn’t made enough revolutions around the Sun to know what we are talking about. :-)
Fifty five revs and counting, been there, seen that. Sleeping cars on trains have nothing to do with this article, because unless your final destination is directly next to a train station, self-driving sleeper cars are going to have a massive advantage over trains, airplanes, and even buses in the future. Especially in the US where we don’t have a high-speed passenger rail service, and likely never will thanks to our geography and our culture.
I’m going to take issue with this.
Having spent a modest percentage of my childhood traveling huge distances (Michigan to the West Coast) on the old passenger system (a magic experience at that age), the argument that a car is better than a train because it takes you exactly where you need to be ignores ride services (including conventional taxis), which have been available since the time of horse-drawn carriages.
Plus you don’t wear your own car out, and the entire scenario ignores weather which isn’t optional in most places not to mention the dismal state of our road system. Yes, ORD/BOS/EWR etc. close occasionally but that’s a matter of hours. Check out what a good blizzard will do to the trans-continental road system in the US this winter.
Further, the supposition that people are willing to endure several days of travel in a road vehicle of any type ignores the reason why people fly: Speed. Let’s start with business travel: nobody’s going to roll out on a Monday morning to meet with a client…maybe on Thursday… if the roads aren’t snowed shut, closed for construction, or underwater.
There are a few sales reps/engineers that do this by automobile, but more than a few actually get involved in civil aviation – flying themselves – if the travel and deadlines get too onerous. We can hope that Zoom et al. will change this a lot, but I wouldn’t hold my breath. Vacations? You have a week. Family emergency/holiday? Similarly. Speed wins, particularly if it’s cheap, even if it’s awful.
People will tolerate being locked up in a pressurized metal tube blowing along at 85% the speed of sound, because it’s cheap, brief, and the reward is immediate.
This may be the epitaph of a clickbait culture, but it’s the one we live in.
I’ve mentally started replacing sentences like “when we all get self-driving cars” with “when we all get flying cars.” It’s pretty effective. Investors or not, totally autonomous cars for general use on all roads ain’t happening any time soon.
Depends on your definition of “soon” and “good enough”.
Musk is predicting fully autonomous vehicles in 2020. His predictions are usually late, but they haven’t been wrong so far so… maybe 2022 will see full autonomy.
And as pointed out on another blog, “good enough” doesn’t have to mean perfect, it only needs to be “better than human”. Once self-driving can be proven to be better than human, there will be enormous pressure from insurance companies to make the feature available.
And as also pointed out, “good enough” doesn’t have to mean “all conditions” either (such as snowstorm driving or backroad country driving). There’s enough completely mapped territory and good weather in the US to sustain lots of business models based on self-driving vehicles. We can still use human drivers for the really hard cases.
Two years doesn’t seem like such a long time to wait.
“Good enough” needs to be _equal_ or better than human for safety. That goal is apparently still a long way away.
Waymo and Cruise have, from what limited documentation is publicly available, accident rates around 2-3 times human. They’re getting hit most of the time rather than vice-versa, and they’ve all been essentially fender-benders. But it looks like their (still) erratic braking and hesitant navigation of intersections is causing trouble.
They’re playing it very conservative, which is probably appropriate at this stage. But it’s hard on the human drivers who share the road with them when they do “crazy” things. It’s like over-cautious teenagers merging onto highways.
We also don’t know what percentage of the time a human override saves a pedestrian/other driver’s life.
Humans are still significantly better drivers. I would guess more than two years on this one. Ford and Toyota say something like 10-20, which puts them on the previously mentioned flying-car timeline. But we’ll see. I’d be stoked to be surprised.
>”“Good enough” needs to be _equal_ or better than human for safety.”
Which human?
1) Most people are better drivers than the statistically average driver, because the average is loaded up with multiple crashes, DUIs, breakdowns etc. that happen to a small minority of really terrible drivers.
2) Even accounting for that by using median instead of mean as the definition for “average”, half the drivers out there would still be safer just driving themselves, and another quarter or so would only see a marginal improvement.
3) The statistics for self-driving vehicles are very fuzzy because they keep changing the software and the cars, and the criteria for safety, and you can always claim that the -previous- model was dangerous while the current model (for which there isn’t enough data yet) is safe. You can’t ban the car either without enough data to prove that it wasn’t safe, so it will take decades in legal battles before anyone admits anything…
4) Corporations will pick the lowest fence to jump, and pull all the tricks to put inadequate and unsafe products on the market before their competition manages the same, or because they need a gimmick to sell their otherwise lackluster or overhyped product. (See, Tesla)
5) Autonomous driving takes away your personal choice to drive safely. Now you’re subject to a machine that is -allowed- to kill you with a certain probability, which you have no influence over whatsoever.
6) Autonomous driving leaves no human being to blame and puts a price tag (the cost of insurance for an automotive company) on human life.
– Personally I hope any companies pushing autonomous cars get sued out of existence on short order. Imaging your spouse was run over while crossing in a crosswalk and killed by a human driver. It is horribly tragic, but do make mistakes. And you might even find a way to forgive them someday. I couldn’t see ever giving the same forgiveness to a machine of ‘oops, I guess the technology isn’t perfect’. I hope anyone injured goes after the company, and while I’m not generally in favor of lawsuits, I’d completely applaud them here. The whole idea is a combination of lazy people and corporate greed, we have no need for self driving cars. Sure, the idea sounds cool, and I like the idea of one myself, until your kid is run over by a machine that was ‘good enough’ for a corner cutting corp. to push to market. And everyone is going to be pushing the envelope of what they can do, while trying to cut costs. Bad news all around. And who programs the ethics of ‘rear end the schoolbus or swerve and take out the mom pushing the stroller?’? Not programming we should be doing in my opinion.
“And the same economic factors that we’ve been insisting will make it far more likely that autonomous vehicles will penetrate the freight delivery market before we see self-driving passenger vehicles are at work with people moving. ”
Your current argument applies to that field as well.
Just want to say: Awesome artwork!!! I’d absolutely travel in suspended animation carried on a back of a robot! (c:
I know, right! I was so excited to see this one when Joe sent it in for the article. I’m honored to have another fantastic Joe Kim piece at the front of my article!
A mate of mine is a 747 captain who “delivers self-loading cargo”.
That was because of invalid urban planning.
Amen!!! Road planning is an engineering problem, unfortunately generally drowned in politics.
No, it was because my parents paid to send me to a private school that was three towns away, and contracted to a bus company that was the low bidder and picked up kids from four separate towns to be delivered to two different Catholic schools, one in the middle of a city. We were lucky enough to be the first ones picked up and the last ones dropped off, and we still had a 15 minute walk home from the bus stop.
I pity you in the USA. Here in europe you can catch a train crossing multiple borders. Or for as short quarter of an hour trip. Most trains run every half hour up till midnight. Some are fast. Some are really fast(ice, talys) and some are slow local trains thar still go around 110km/h.
Here in regional Australia, some towns have a daily bus going through. Some towns have a daily train. Some towns have a daily flight to the nearest capitol. Some have a mix. Some have nothing and you have to drive to another town to get any kind of public transport. The roads have hazards that jump several metres at once (kangaroos), as well as things like potholes, oversized agricultural vehicles, mobs of cattle walking, water across road, roadworks. Problems at 110km/h sound much scarier than incidents at urban speeds.
Frequency of flights is only an advantage if you have lots of flights, which you only have where lots of people use them. If there is a better option, it needs to hit critical mass to be able to scale up and then be way better. Chicken and egg problem. Outside the cities and motorways, humans will be driving for a while yet.
“All forms of regional transport curently suck in Australia” got it. Did you have a point about self driving cars?
I’m in Aus and do – until they can recognize a kangaroo at the side of the road that might leap out, I wouldn’t be using them in country areas.. Until they can do that better than me, nope.
And before you all laugh, this is a real problem. On my last long drive – last week – I do 1000km each day for two days (as driving is way better than planes for a lot of trips in Aus) and twice on that single drive I ended up on the wrong side of the road at 110km/h (or so) to miss a kangaroo..
But the bigger problem with autonomous vehicles on the road is legal liability. I suspect that is going to kill the concept in most countries for a long time, it doesn’t even matter when they will be able to drive better than a human…
I’d prefer hitting a 90kg Roo over a 300kg Moose.
true – I’ve had to miss moose when I driven in Canada – but moose are much bigger and much slower, they don’t leap out in front of you from the bush at the side of the road…
With respect, I pity you in Europe, and for that matter anyone who hasn’t driven cross-country in North America. Unless and until you do, there’s no way to wrap your head around the concept of vastness. I grew up in New England and spent most of my life accepting the fact that driving anywhere was a few minutes of city driving, a few minutes crossing the suburbs, and a few minutes in the country. The cycle would then repeat itself, and even if you were “in the middle of nowhere”, you knew that it was just a few minutes before you’d find a store or a gas station.
Then I drove across the US. Coast to coast (almost). I would never have believed the scale of this country if I hadn’t done that. Most of the states west of the Mississippi are bigger than many European countries. Hell, Elko County, Nevada is bigger than Switzerland! Until you have boots – or tires – on the ground here, you’ll never understand what the real meaning of “the middle of nowhere is.”
Bigger is not always better. :) I for one hate flying because of all the poohaa at the gates. In between its just a crampy bus with wings and an attitude. Self driving cars are in europe a solution looking for a problem i think. On long roadtrips in nevada they may come handy though.
https://upload.wikimedia.org/wikipedia/commons/b/b5/SNCF_T2_Perpignan.jpg
I was lucky to use wagon lits. The most big advantage, compared to take a plane was that railway station are downtown. The other one was the classic an a bit decadent interior. The rytmic noise of the expansion joints were an aid to get the sleep. By the way heaw fog wasn’t never a problem due the signal repetition system.
Overnight sleeper trains and overnight sleeper buses are common throughout Asia. They can be an effective way to get someplace. Not only do you save the cost of a hotel room, but you also save the time one would have spent traveling during the day. While it is possible to sleep in transit, this is typically a poor quality sleep. There are lots of factors that go into success of the services. All mass transit works better where there is a high density of people. Road quality, personal safety, cultural acceptance, food availability, availability of onward transit, amount of drama, and income of passengers all play a part. Personal security and security of personal items which can disappear when sleeping play a role. On one sleeper bus I took in China we arrived at our destination at 4:00 in the morning. The bus, allowed anyone who wanted to get off to get off, but then then the driver and 90% of the passengers slept on the bus for another two hours enjoying more restful sleep and waiting for first light and local transport to be available. Many Americans already drive great distances in cars, and many have carpooled with friends enjoying the benefits of trading off driving. No doubt self-driving vehicles could make this better.
Everyone saying “But trains…” is missing the point. Trains, buses, and airplanes all have the same problem. You have to go to where they leave from. You have to wait to get on the transport. You leave on their schedule.
When you get to another of the limited number of places trains, planes, and buses travel to and from, you have to rent a vehicle and drive yourself or hire a cab and get driven to your destination. In some places you may be able to take a local bus to get somewhat closer. From there you’ll either have to walk or take a cab or rent a car from a company that brings your rental to you.
An autonomous road vehicle could *come to you* then take you right to the front door of where you want to go. No long lines, no waiting on the schedule of the transport. No extra expense on taxis, buses, parking your car in a lot where it can get vandalized. If you’re running late getting ready to leave, no problem. The vehicle isn’t waiting on anyone else.
Mid sized autonomous road vehicles with bunks could run routes with pickup and drop off of passengers at various stops, or they could be hired for non-stop runs.
The ‘killer app’ of these would be that convenience and flexibility of use, which would save passengers a ton of time and money.
For trips of 200 miles or less it’s possible to beat a plane with a car and two drivers to switch, thanks to the long delays just to get on the plane. It’s real easy to beat a bus with a car, especially on routes where the bus stops at every little rinky-dink town along the way.
EXACTLY!
A friend of mine has a great idea for dealing with the “no car when you arrive” problem. Take your car with you!
He designed the “Tango” — a 2-passenger car that is only 3 feet wide and 8 feet long. They can “lane split” in most states in the US, doubling the capacity of highways. Four can park in a single parking space, for up to 4x more parking spaces. See http://www.commutercars.com
The length was chosen so they fit crosswise on a train flat car. Put ramps on the sides of the train car. Train pulls into station, you drive on, and have your own private “cabin” for the trip. When the train reaches your station, you drive off. Now you have your own car to drive to your destination.
This, or some other alternate future could change the way we travel. But the status quo is powerful — we’d rather spend fortunes to support old solutions, than to look for new solutions that may well be better and cheaper.
Each Tango is >1 ton, 4 is putting a lot of weight into 1 parking spot.
We already have SUVs in which the rear section converts between seating and cargo. How about one that converts to a bed(s)? I wouldn’t be for everyone but it would be good for some people.
Alternatively, how about seats that (fully?) recline and have amenities like first/business class airline seats? Keep ’em upright for daily driving around, put ’em down when you want to sleep. Maybe even swivel the front seats around (like in a motorhome) so four passengers can socialize, play games, eat, etc.
I think there are multiple ways that a family car, or family second car, could function well for overnight sleeping trips.
Soylent freight is people!
(somebody had to)
Would make more sense to have the bunks on the lower level, and the lounge on the upper level. That why there should be a steadier ride for sleeping, and less likely to get tossed out that ungated bunk opening.