Hopefully within the next 30 years or so, we will gradually phase out driver's licenses. Much like how smoking was ubiquitous in the 50s & 60s, but as we as a society came to realize its grave dangers, it gradually became illegal in many public places.
The idea that we allow almost anyone 16 or older to control several thousand pounds of metal that can travel at high speeds really is quite absurd when you think about it. There are more than 30,000 deaths every year due to car accidents, and hundreds of thousands more injuries.
It's amazing how screwed up our society's priorities are. That's ten 9/11's worth of deaths every year from car accidents. By the time the wars in Afghanistan and Iraq have ended, we'll have spent around $4 trillion on them altogether[0]. If even 1/100th of that had been redirected to research on self-driving car technology, we might have already saved tens of thousands of lives.
> Consider it rationally. Driving is costly, but it engenders an entire host of benefits & pleasures I don't think you're even considering.
I am considering it entirely rationally. I believe that we should be investing more money into research that could save 30,000 lives every year. Never did I suggest in my initial post that human driving should be banned before self-driving cars are capable of replacing humans in this capacity. If done correctly, self-driving cars will have all the benefits of current automobiles, and then some.
As for the pleasures of driving - that's what a whole lot of people said about cigarettes, but look where we are now. And don't be silly - no one's personal "pleasures" override the safety of other people's lives.
Research? We've known for centuries how to get by without conveying ourselves everywhere in 4,000 pound steel cages. It's called population density, mass transit when a community exceeds a certain size, and having vital services (drugstores, groceries, maybe the odd pub) within walking distance of where people live. It's not a technical problem, it's a social problem.
it engenders an entire host of benefits & pleasures
what is this host of pleasures? I get that some people enjoy driving. I loved it when I was a teenager, now I hate it. Most of my driving is done in traffic and I'd much rather read a book or take a nap insted of micromanaging my 20km/hr commute.
Not only would autonomous cars give me back my 1.5hrs/day that is dead time... but the roads would be incredibly more safe.
And for people who like to drive... how about a racetrack?
The greatest pleasure I get from the roads - on my scooter :) - is from other traffic. Racetracks are a different kind of fun, for when you want to go fast and corner hard.
I get that you get pleasure from driving or riding and if it's anything like my experience (in my teens) a good part of it is the danger factor. I drove like a lunatic when I was 16 (just like most other 16 year olds) and it was great fun at the time. Now when I look back I can see what kind of an idiot I was for putting others and myself in that kind of mortal danger.
That's not to say that you can't enjoy driving responsibly. But there's so much unnecessary danger and tedium involved that it seems a no-brainer to move to a computer controlled future.
It's actually less danger factor than it is the pleasure of competence, doing something well and smoothly. I live in London, where there are a lot of powered two-wheelers, filtering is legal, and it's just about the best way of getting from A to B in the congested city. Sure, it gets slightly more dangerous if you get into a bit of a race with someone else - the adrenaline of competition can push you a bit, but you're mostly just a risk to yourself.
But being able to skip past long queues of stopped or slowly moving traffic, safely and efficiently almost as if the traffic didn't exist, feels really good.
Honestly, I think inner city scooters don't really factor into this debate. As you mention you pose a limited risk to others and your choice to risk your own well being is exactly that- your choice.
2000kg passenger cars travelling at 60-100km/hr are another story altogether. My guess is that there would be a tiny fraction (<1%) of all road fatalities caused by purely mechanical failure. That leaves 99% (stats pulled from my ass) caused, or largely contributed to, by human error. For a large class of these errors a computer would never cause the problem or safely navigate around it.
Saying "I want to keep driving 'cos its fun!!" isn't really good enough in that context.
FWIW, it's not really an inner-city scooter - at 300cc, top speed about 140km/h, it's well capable of touring, and I have toured on it.
But even if it was, it seems to me that you're assuming a boundary where none exists. There is no fixed point in the road where it's suddenly urban vs suburban vs rural; the most marked points are entry and exit to motorways, but they have lower per-mile risks already.
I'd be the first guy to say that 2000+kg cars ought to be move more safely and predictably; I'm a committed biker, and that would be safer for me too. But in the chaos of UK traffic, I don't really see how it's workable without converting everything - absolutely everything, pretty much overnight - to automated vehicles.
I mean the streets have one lane most of the time, but occasionally two or more lanes between certain traffic lights, junctions are not regular, roadworks at some point in your journey are statistically almost certain, pedestrians are always wandering out into the road, wobbly cyclists, etc. The only way I can see an automated vehicle working in that environment is one that moves unacceptably slowly to mitigate the risk of tort suits against the manufacturer; or an entirely separate road network, strictly for automated vehicles.
It would be interesting to know the injury/fatality stats for the kind of roads you're describing. If the rates are high (relatively speaking) then I would argue that yes a total conversion to computer controlled driving (when available) is the only socially acceptable thing to do. Public safety would trump your personal enjoyment of the act of riding.
Realistically speaking though truly autonomous driving in tricky (even for humans) conditions like you describe is probably a long way off, although within our lifetimes I would think. The other option is a total rethink of the city/suburb nightmare that we now live in. But that's a debate for another time :)
Um, you must be joking. Frame it like an opponent would who is running for political office against you in the year 2050. So you're saying take away people's privilege to operate a vehicle they own and paid for with their own money? You want to require people to use an automated machine that, in theory, the state could over-ride the controls of at any moment? You want to eliminate all jobs associated with commercial driving (buses, delivery trucks, etc.) many of which are union jobs?
No. No. No.
Also, the answer to the danger & death problem is well understood -- you simply require people to first accumulate several dozen hours behind the wheel, under supervised instruction, with training for emergency situations, just like we do now with pilots licenses. The fact that this hasn't happened yet in the US (like it has in say, Germany) should show you what a political impossibility what you're saying really is.
P.S. For added hilarity, the story directly below this one on HN is about cell phone tracking, where the top post is all talk of going without cell phones because the device's tracking could infringe on people's freedom.
Which is the scarier dystopia? Your phone keeps a running tab on your location, which could be accessed by the state or Google? Or your mode of transportation is tracking you, _and can be over-ridden or disabled at any time by the state or by Google_...
I'm FAR more frightened of being t-boned by a drunk driver than of big brother taking control of my car. Honestly can't tell if you are trolling or serious.
I was seriously trolling, also known as taking an idea to its logical extreme to make a point.
You're presenting a false choice here. There's lots of other ways to attack the drunk driving problem without taking everyone's drivers licenses away and banning normal driving, as the OP was advocating.
Ok I see what you were getting at now but I still don't agree with your conclusions and here's why:
1. Anti-drink driving campaigns and counter measures don't work. People make the choice to drive while drunk when their decision making powers are at their worst.
2. The issue with human controlled driving goes far deeper than just DUI. The fact is that we're (collectively) just not very good at it. Even if you are the best, most cautious most defensive driver there's nothing you can do about some idiot running a red light because he was adjusting his stereo. Your perfect driving record is intact but you're dead anyway.
3. The chance of being injured or killed while driving are astronomically higher than the chance of big brother having the motive/inclination to want to remotely control your car (although I might feel differently if I lived in Syria - even then this is getting really close to tinfoil hat territory).
The way I see it computer controlled cars shouldn't be seen as curtailing freedoms, rather as relieving us from tedium and making us safer.
I don't think your point 1 is as true as you think it is. Anti-drink driving campaigns seem to work: http://www.thecommunityguide.org/mvoi/massmedia_ajpm.pdf describes studies that identified significant (>15%) reductions in alcohol related accidents, with economic payoffs during the campaigns of more than 20x.
That's certainly a LOT better than I would have thought, although still depressingly short of the 100% reduction that computer controlled cars would give us.
I doubt we'd see a 100% reduction of accident injuries and fatalities with automated control of cars - diminishing returns due to other uncontrolled factors like pedestrians, dropped loads, pathological algorithm responses in unusual situations etc. Of course, spending 6x as much on road-safety campaigns as in the referenced studies won't deliver anything like a 6x improvement either.
It's easy to think of the political obstacles if you are expecting this to happen overnight. But it will be gradual. People will share the road with robots for a while, and when it's clear how much safer the robots are, there will start to be roads and then areas where people aren't allowed to drive.
As for "tak[ing] away people's privilege to operate a vehicle they own and paid for," they can feel free to drive on private property. Even today I'm not allowed to drive my off-road truck on public roads because it doesn't have proper smog equipment and mudflaps, and I own and paid for it with my own money. It's really not that unreasonable when it's a matter of public safety and the restrictions only apply to public roadways.
As for the automation stealing jobs argument, that's a whole 'nother can of worms.
try to take away my motorcycle and i'll buy a gun.
seriously, what's wrong with you? individual mobility is a great freedom and source of happiness. i like to be able to control a machine, drive it up a mountain. i've spent months driving all across the US, Europe and places like Iceland. best experiences ever, every single turn was a choice by me.
the minute you make every thing super safe people will kill themselves out of boredom.
but i guess you're one of those people that want to ban drugs, sugar, fat, knifes, flight, compilers and anything with sharp edges.
and your numbers, oh scary, TEN 9/11s. maybe all that shows us that on a historic scale, 9/11 wasn't that big? how many people got wiped out on D-Day? By the Spanish Flu?
It seems the most knee-jerk reactions always come from bike riders (buy a gun.. really?) :)
How about the upside. Every bike rider I've ever talked to bitches about stupid "cage drivers". Imagine if every 4 wheeler was autonomous.. no more bitching, you'd be able to zig-zag between them easily, and if you do happen to lose concentration for a moment you can rest assured that a computer will (probably) stop before it kills you.
the minute you make every thing super safe people will kill themselves out of boredom.
Really? You think people would kill themselves because they no longer had to manually do one of the most boring, repetitive, tedious everyday things they have to do? Personally I'd read a book.
Personally, I'm all in favor of having (potentially dangerous) outlets for people to have fun controlling machines and driving them up mountains. I would just rather those outlets were separate from, say, people's means of getting to work in the morning.
This is ridiculous. You're seriously suggesting that you're going to shoot your way to mobility :) I'm assuming that's just some macho bullshit because it's insane.
What you're not realizing is that it's often not the person causing the accident that gets hurt, the idea of making something as tedious as driving automated is that it's safer for everyone.
Drugs already are banned. The ones you're most likely talking about are anyway. And the reason is because they fuck people up.
How about people do something productive with their lives instead of taking drugs, developing diabetes from sugar, overweight from fat, and kill people with large heavy steel motorized machines?
You can try to belittle 30,000 people but if you stop cleaning your guns to think about it long enough you might actually appreciate that all those people probably didn't want to die from theirs and other people's driving mistakes.
and yes, i believe in every adults right to fuck him or herself up. if i want to exercise, let me. if i enjoy fatty food, let me. if i enjoy driving my car, let me.
and why do i need to be productive? who says that? life is short, why be "productive" all the time? how about some fun, self-indulgence and mindless games?
we already have rules to protect the general population, way too many in certain areas, far too less in others. speeding is forbidden, pot as well - but areas like finance are wide open.
how about focusing the efforts of banning things on stuff that impact millions, at the same time rather than stuff that will hurt me and maybe a few folks around me? is that line of thinking too European? does everything in the US that might lead to unproductive, un-puritan fun be banned? nudity, foul language,...how dare people do that?
Well first off, you're talking to a European, and while I'm sympathetic to the fear of the encroaching 'nanny state', I just don't share your solution (guns)
If you want to get yourself fucked up then you're most likely quite happy to have the rest of society bear the burden of your subsequent health care, ditto for your diabetes care etc. I'm sure you get what I'm trying to say.
None of us live on an island, our actions impact others and when it does so (as Spock said) the needs of the many outweigh the needs if the few.
This country (America) is already being financially crippled by health care costs so reducing sugar and fat and car accidents seems to make perfect sense, even at the cost of a few's peoples' adrenaline fixes.
I don't disagree with you on Spanish flu (50m people died), or on the corrupt financial system, but that's a fragile house of cards. Economics is a complex system and it's all based on a mutual agreement in the value of little pieces of paper. It's a social contract that is holding together the fabric of capitalism. While that night not be the perfect system, it's so far the least bad that we've found.
"why do you need to be productive" - maybe I came across as being a little self righteous (which is far from the truth), I was just trying to say that one persons fun is very often a ton is work for the rest of society, those that scrape up the parts after a motorcycle accident, those that provide health care and those of us (all of us) that pay for it in our insurance premiums. Why should I, someone who is fit and healthy pay for everyone else's indulgence in fat and sugars that bought about heart disease and diabetes. The answer of course is that same one that is the reason why fixing cars safety is important, because the individual sometimes has to sacrifice something for the success of the rest of society.
Oh, and as for priorities: the cost of eliminating danger from traffic to where it is "trivial" or "absent" probably isn't financially viable. Meaning you'd have to spend an inordinate amount of money for ever slimmer returns.
If you follow how rules and regulations aimed at making traffic more secure over time you'll notice that the incremental improvements are becoming smaller and smaller. In most sensible jurisdictions, the important, big stuff that saves lots of lives is in place and people are now bickering over the little stuff that is perhaps going to same a small number of people. Or, in many cases, won't save a single person, might even cause more accidents, but generates more revenue.
If you want to analyze something in terms of value to society you'll have to be prepared for a lot of unpleasant aspects.
That's how we improve the society: not by improving individuals, but taking away responsibilities from them. What comes next? People destroy their health eating crap, so we just plug them into feeding tubes?
There are many benefits to self-driving cars, and sure, security may be one of them, but don't delude yourself thinking you can solve all societies chores by restraining what people can do. People still smoke, even though it's known to be harmful. People do stupid sh. Neither technology or regulation are substitutes to wisdom.
Except we won't. Because flying cars is called "aviation" and aviation brings new dimensions of "hard" in addition to a third spatial dimension. And truly self driving cars that require no dynamic input from humans at any time is something we'll get asymptotically closer to, but will never quite achieve except in relatively specialized cases.
This is what I believe. I can't prove it, but it is what my gut feeling tells me.
And why would you want to? Of course, for typical highways it is obvious that it would probably be both faster and safer to have some sort of mandatory cooperative autonomous mode, but even that will be hard to achieve. I've met some of the people who work on developing and standardizing certain inter-car protocols to solve these types of problems. Most of the people I've met are neither brilliant nor capable. For the most part they illustrate to me a) how hard the problem is at scale (ie. get all manufacturers onboard) b) there will be a lot of idiots involved in this which means it'll take an awful lot of time to even come up with even baseline systems for proper crash-avoidance -- nevermind the real-time dynamic systems for controlling masses of fast-moving vehicles.
Some benefits of self-driving cars only come in when they hit certain market-shares. For highway driving there are big gains when you have enough for a dedicated lane, and when cars can "draft" one another, but the biggest gain will be in cities when (if) we get to 100%.
At that point we can do away with all traffic lights except for pedestrian crossings. Not only will stop signs be unnecessary, cars could approach intersections at speed, knowing that there was a gap in the cross-traffic for them to fit through. If you let a single human-driven car onto those streets you couldn't implement anything like that, though.
The flying cars argument is silly. Yes it was a fanciful 1950s dream that never made it to reality... however autonomous flight is real.. a good portion of commercial air flight is done under autopilot (including landings).
We never even got close to flying car tech... but saying that truly self driving cars that require no dynamic input from humans at any time is something we'll get asymptotically closer to, but will never quite achieve is silly, we're already very close to this reality. I'd be very surprised if in 10 years we don't have a completely computer controlled car.
Why is it silly? Consider this: we've had autopilots for many, may years. But even very limited gadgetry such as automatic parking is a very recent development in cars. And that is an almost trivial scenario.
(The reason the flying car won't be a reality is that all of those (auto-piloted) air planes do in fact have a crew of 1-3 highly qualified professionals -- and the flying cars would be piloted by the sort of people who can't figure out roundabouts. Also, the density of planes is extremely low compared to what would be the situation if flying cars were commonplace).
Also, planes have maintainance regimes that are vastly different from your car. I bet that you did not go through a checklist when you got into your car this morning, nor when you parked it. You probably do not have the faintest idea how the fluids are doing. You probably don't even know if all the external lights work. It is highly unlikely that you even performed the most rudimentary tests of all before getting into your car: walking around it and kicking the tyres. There was no qualified and certified mechanic (for your specific model) to go over your car prior to setting off.
All of this is true.. however you categorically stated that we "will never quite achieve" totally autonomous cars. To me that sounds like a silly statement especially when compared to flying cars (a totally different class of problem).
The fact is that we're very quickly edging towards fully computer controlled cars. You have to have a truly limited imagination to think we will never get there.
The argument about airplane maintenance also doesn't hold water. Result of a malfunctioning plane: plunge 30,000ft and hundreds dead. Result of a malfunctioning car: coast to a gentle stop. See any difference?
I was being imprecise. Sorry about that. I don't think that we will see totally autonomous cars _in_our_lifetime. "Never" is not a useful timescale.
You have to have a truly limited imagination to think we
will never get there.
40 years ago people did the same sorts of extrapolations on air-travel and figured that 15-20 years later all commercial air travel would be supersonic. And why wouldn't they? Orville Wright was still alive when the first supersonic flight took place. We went from not being able to fly heavier-than-air machines to breaking the sound barrier in 43 years. But of course, that didn't happen.
And in just the past few years we've gone backwards. There is currently no supersonic passenger jet in service.
A lot of domains appear to exhibit the same asymptotic behavior. Space travel, medicine, AI etc. Domains where initial speed causes undue optimism and where we later suffer regressions. (No regressions in medicine you say? Well, how about the crazy anti-vaccination people and the re-emergence of diseases that were practically eradicated?)
I build stuff, I program stuff, I occasionally build and race cars. I don't think my skepticism stems from _lack_ of imagination -- I think it has more to do with an abundance of imagination.
The argument about airplane maintenance also doesn't hold water. Result of a
malfunctioning plane: plunge 30,000ft and hundreds dead. Result of a
malfunctioning car: coast to a gentle stop. See any difference?
I'm sorry, but that is pure nonsense and many, many, many people pay the ultimate price to prove you wrong every day.
I don't think that we will see totally autonomous cars _in_our_lifetime
Now you're moving the goalposts to suit your argument.
Supersonic flight was abandoned because it is expensive and prohibitively noisy. Will some set of obstacles also arise to stymie autonomous vehicles? Maybe, but we wont know till we try.
The regression in medicine is due to pure stupidity. Will the same or similar type of stupidity raise its head to block self-driven cars? Again the answer is maybe. Maybe not. We wont know till we've tried.
many, many, many people pay the ultimate price to prove you wrong every day
You misunderstood what I said. My point was that if a computer controlled car's computer fails it can be made to fail-safe. Shut down and coast to a stop. The same cant be said for planes.
Of course we will have scenarios of computers being mis-programmed and causing death. The point being that once a bug is found and eliminated it wont reoccur. Not to mention that a human will never drive as well or recover from trouble (such as a fishtail) as well as a computer. Our interfaces are simply too slow, clumsy, prone to fatigue, prone to mis judgement due to adrenaline etc. the list goes on.
I get what you're saying about optimistic extrapolation being a trap... but we don't have to extrapolate far from where we are today. The google self driven cars have already driven tens of thousands of miles with the only recorded accident occurring when the vehicle was under human control.
Not only that but Nevada has already licensed the cars for testing on its roads. Now you can argue that the testing is not 100% authentic. From what I understand the routes are pre-programmed and 2 occupants must be in the car at all times. But to say that this tech wont go anywhere for the next 50-60 years (however long you intend to live :) ) ... seems to lack imagination.
This isn't everyone using personal jet-packs or interstellar travel we're talking about here.
> You misunderstood what I said. My point was that if a
> computer controlled car's computer fails it can be made to
> fail-safe. Shut down and coast to a stop. The same cant be
> said for planes.
Just because you are at ground-level you are not safe if you lose control of the vehicle.
> The point being that once a bug is found and eliminated it
> wont reoccur.
We already have a software industry and we know that this doesn't happen: bugs do not occur only once and then get eliminated forever. There are people making a living writing books about mistakes that people repeat over and over in software.
I've come to realize that autonomous cars are a pre-req for flying cars. Obvious now that I see it, of course. Humans as a whole can't possibly handle flying machines when they have so much trouble with cars.
I'm skeptical about totally autonomous cars, but I think freeway driving will be increasingly automated. Thrun has said that he's let his car drive him from the bay area to Reno several times. I assume he's exaggerating a bit, but probably not by much. It's a solvable problem for good road conditions.
There is no way a mass market for personal air vehicles can happen if we need traditional pilot's licences. Following the eventual success of driverless cars, one can imagine that all-weather PAVs will not require a license because we will just be passengers.
Your argument allows for certain driverless cars in which a human is present to provide dynamic input for situations that require it. This is currently the legal requirement in Nevada.
Admitting that driverless cars cannot be truely perfected, does not preclude mass adoption - because they merely need to be safer and cheaper to operate than the current human-only traffic situation. I think we are less than a decade away from that tipping point. Does anyone else agree?
I would be careful to conclude without data. For instance insurance companies have data that suggests that systems for warning the driver of lane drift make cars significantly more likely to have accidents. I can't remember if they had figured out why, but this is the sort of counter-intuitive results you have to expect when you are dealing with a field that isn't adequately understood.
And large populations of autonomous cars in obviously something we have zero experience with so we can't really know if there will be unexpected dominant factors.
I'm not sure they'll be phased out entirely. What happens when all electrical systems go out. What do we do about inclement weather, system malfunctions. What do we do about classic cars (as in people like owning cars from decades ago). What about driving on unimproved roads (in the country). I think in urban areas and developed areas this is feasible, but I think there are some scenarios where it may not happen for a long time.
Why don't we now have self-driving trains/subways? They move on rails but we keep insisting they must be supervised by people in the vehicle (with a few exceptions like people movers at airports and amusement parks, etc).
The DLR is slow, very slow, and services are fairly spaced out. I want automation for safety, speed and much higher capacity used on the infrastructure.
At least one Paris metro line (ligne 1?) has been automated, and it's faster and more frequent than most of the other lines there - at least that's my anecdotal experience from two weeks ago.
> What happens when all electrical systems go out. What do we do about inclement weather, system malfunctions.
You could ask the same questions about human-driven cars. When the stoplights don't work and there's ice over all the roads, sensible people stay off the roads, and people with a death wish go out for a spin.
Wait, what? No stoplights and ice on the road - that's like 1/3 of the year here in Finland. All you have to do is adjust to the environment and you are able to get from a place to another.
Damn this European smugness gets on my nerves so much for some reason.
If you have these driving conditions 1/3 of the year 1) Finnish drivers grow up learning how to handle it because they have to, 2) most if not all of your cars have 4 wheel drive and ice-capable tires.
That is not the case here. Those driving conditions are rare.
In a city, with 10-20% grade practically everywhere? Yeah it's passable in flatlands or sparsely populated areas, but frankly automatic cars aren't really going to be any worse than humans at traversing bad conditions, so an impassable road for an automatic car would be impassable also for a human driver.
I've always thought a simple way to improve the process would be to require people to volunteer 200 hours in an Accident and Emergency ward at their local hospital.
force us to spend time with people who found out what thousands of pounds of metal does to the human body when it decelerates quickly.
I think there will be two nice outcomes:
1. Lots of people will chose not to get a license (and will probably avoid cars altogether).
2. The people that do get licenses will drive with a lot more care and respect.
I've worked in an Emergency Room. Literally every single person who worked there drove a car. Both trauma surgeons were motorcycle enthusiasts. I'm friends with pulmonologists who smoke, psychiatrists who drink, endocrinologists who are fat.
Having seen the possible consequences of a bad decision first-hand does not, in fact, stop people from making bad decisions.
I'm not talking in hypotheticals here. I'm saying that of the many, many people I know who do in fact work closely with the badly injured victims of traffic accidents every day, not a single one has been scared away from driving.
Alternatively, we could require them to ride along with accident investigators on the police force, because police officers are well known for their careful, respectful driving.
accident investigators != police officers behind the wheel. You probably know this, but still had to make an awesome comment about it?
It's funny how some people really think that self-driving cars are a bad thing, automation is a bad thing and people should be kept in labor. Yet, the same excact people, "optimize" their workflows by automating things, writing scripts, making the machines do as much as possible because there's no human-errors, it's cheap and it's very fast.
I really don't see (consumer)cars having no steering wheels, pedals or any other conrol equipment. What I see is cars gradually(within the next 5-10 years and onwards) get more and more of technologies to assist and aid with driving, ultimately resulting in a vehicle to which you can just say the address, and it will get you there. And yes, you'll still need a driving license for driving such a vehicle for decades, even if you don't expicitly take any control over the driving process.
For some it's an awesome future, for some it's a nightmare.
Most delays and jams and other various issues drivers deal with are a result of human error. Suppose a driver on a highway slows down because {they see something interesting, their kid acts up, life happens}. This makes the car immediately behind them slow down. And the car behind that slows down. It's been shown that this forms a wave that propagates backwards at 12 mph or so. The fact is, this delay is caused by a human mistake/issue.
In a world in which every car drives "optimally", or at least without many of the human error mistakes that currently are made, many traffic jams will be a thing of the past. Many issues will cease to exist any any issues that do exist will be ameliorated and fixed by a machine that can do it better than most humans (or at least, the average joe) can. Perhaps one will be able to drive at the speed limit the entire distance from source to sink if the traffic grid is integrated into the network.
The transition to the post-driver road, however, will be difficult.
There was a fun blog post several years ago, which has stuck with me ever since, and I think about it every time I'm stuck in a traffic jam. It was a one-man experiment in "fixing" traffic jams, by willfully slowing to the point where he no longer had to stop/start and could maintain a steady speed. He found that traffic behind him would "unkink", making the drive for everyone behind him a bit more pleasant/safe/consistent. He also found that many drivers would become angry at him, because he was going slower than the cars around him...but the cars around him were speeding up only to have to stop soon after to wait for traffic. Many people are, frankly, too stupid to be in control of a couple thousand pounds of rolling death...but, it'll be a challenge to get people to give up that control.
Hopefully, the ability to read, play games, watch TV, etc. while being driven to work will be sufficient to make people willing to let their smart cars take over the roads. As you note, traffic jams will become a thing of the past in most cases; self-driving cars can also factor in traffic data, weather information, etc. in ways that a human driver probably can't easily/effectively do, making them safer and more efficient in a lot of other regards.
That's right. What you describe has been automated as an adaptive cruise control.
Almost a decade ago, traffic studies demonstrated that if only 20% of cars used adaptive cruise control, then traffic jams would be greatly diminished at then-peak carrying capacity. Or more cars could be carried on a particular road before traffic jams occurred.
Adaptive Cruise Control is currently a feature on many luxury cars, but will trickle down to mass market cars during the transition to driverless vehicles.
A similar effect is in play with 'metering light' systems - total
throughput may be increased (counter-intuitively) by reducing
localized traffic densities and smoothing irregularities in flow.
But people tend to perceive speed more readily than time to destination,
and tend not to be very aware of total traffic throughput at all.
When you go to driverless cars you can have higher traffic volumes on the road before you have to slow down. However the fundamental mathematical fact behind car jams is that you can pass more cars/second at high speed than at low. So if you have a road running near capacity at high speed and anything goes wrong, then you get a car jam that by necessity must travel backwards and grow in the process.
During peak commute hours, the volume of people who want to be on the road exceeds their capacity. As long as this remains the case, we'll continue to have car jams.
The idea is that automated driving, when it's good enough, increases road capacity enough that jams no longer occur. Capacity improves because all cars are now able to go much faster and with smaller distances between them, and without constant mistakes slowing other cars down.
Jams are like a constant. If there are no jams, people can live further and further from their places of work and play without paying commute penalties, so they'll buy larger properties further and further out, until the congestion from the extra commuting becomes an issue again.
The problem is that when you increase capacity, and decrease flow problems, you increase demand for that capacity. I find it very hard to believe that you can increase capacity enough to avoid jams in cities like Los Angeles.
Ok, interesting business opportunity, time-based road pricing auctions integrated with automation. The roads keep flowing because people will pay others to defer their travel.
This is my thought, too. I think self-driving cars are extremely cool, but I think that a lot .
Once self-driving cars are "the new normal", they'll be a lot more comfortable than regular cars. Which means that a lot of people won't think twice about a 75-minute car commute, because with internet access and the ability to read and an extremely comfortable chair, it won't be unpleasant to do so. And we will (because our reaction time is slow compared to a computer's) be able to raise the speed limits once self-driving cars are the norm: 150 mph highways where human-driven cars are banned, just as you can't take a bicycle on 65-mph highway. None of this is inherently bad, but I hope we won't be burning hydrocarbons to do it.
Whenever I'm in a traffic jam (expressway), it is usually caused by a lane closure. There are almost always signs posted alerting everyone of the closure, usually with more than enough distance from the problem area that no one should have difficulty getting over to an open lane.
What seems to me to be the problem is people follow a greedy algorithm. They cut into the closing lane because it's empty (or sparsely populated), then attempt to jet to the forced merge point then merge. At some point too many people do this, then you get two lines, one in the closing lane, one in the open lane that have to come to a complete stop and take turns.
Remove people's ability to follow a greedy algorithm and I predict you'll see a dramatic decrease in traffic jams due to lane closure.
>In a world in which every car drives "optimally", or at least without many of the human error mistakes that currently are made, many traffic jams will be a thing of the past.
Jams are not only capacity problem, but also sociological. People are willing to tolerate certain waiting time to get to fun/profitable location. If there are less jams and roads are free and gasoline is cheap, people will drive more.
While they may obey the speed limit, they will also drive in the right lane so it's OK. What is annoying are people blocking passing lanes which results in aggressive drivers making less than good decisions about maneuvering around the blockage.
Its not really important what speed driverless cars actually settle on. What matters is that they are consistent. The most dangerous thing about human drivers is that they are erratic.
I would not be at all surprised in the coming world of robo-cars to find that humans who still wanted to drive themselves will need special licensing and training, much like pilots of today, if nothing more than to learn to be predictable.
Actually it is very important from a fuel economy perspective. Air resistance is quadratic in velocity, so a car that gets 35mpg at 65mph will get, say, 20mpg going 85. On the other hand, the possibility of convoy driving (http://phys.org/news/2011-05-drivers.html) means you might be able to have your cake and eat it too: rapid travel with little efficiency loss from wind resistance.
I'm not sure I'd feel safe about convoy driving unless there were some kind of way to manage emergencies (like a tire blowout, engine problem, deer) which under normal circumstances isn't very dangerous because you have a second or two to manoeuver. But in a close convoy system, there is little room for error and even an automated system may be unable to compensate for the emergency in a timely fashion thus resulting in a chain reaction crash --the kind we see in foggy conditions with human drivers.
One option to reduce convoy risk is have cars attach to one another such that no single car can cause a failure in the overall system (similar to trains).
We shouldn't think of automated cars as independent as we do. They are in fact simply compartments.
With that in mind, imagine cars attaching to each other to form trains and break apart as they begin to reach the desired destination.
This also has impacts on fuel efficiency since drag can be reduced across the entire system.
(Cars don't have to be the same model, they just need a formalized attachment system on the front / rear).
Yeah, I think buffers (like in trains) could be an option, maybe magnetic. If it requires physical contact, though, that would mean a whole train of cars (maybe cars would be grouped into sets of ten or so?) would have to speed up and down as cars joined and left the caravan --I know we do this in everyday driving, but it's "softer".
[edit, just thinking, if we knew destination in advance, the first car would be the one going furthest and the last car in the group going the shortest so that the interruption would be inconsequential]
Back to accidents, there will the edge cases. And because it's programmed (and not a human "accident") people might find more fault in the system than they would in a human. So even though accidents and deaths would/could be reduced considerably, those that would occur might be seen as more negligent than accident. Just thinking out loud.
I'm sure that convoy driving is another 15-20 years in the future from the point where self-driving takes off. However it's not like those dangers you mention cannot be managed. Self-driving cars already require a wide array of sensors - together in a convoy you get such an amount of data that the computers should be able to get out of hairy situation at any point.
I believe Google has publicly said a few things that are slightly contrary to this. Their vehicles are allowed to exceed the speed limit by 5%-10% in order to keep pace with a car in front of them. They do this because currently there are few things as reliable in terms of road information as another car being driven by a human. So the code makes an exception basically saying that its better to break the speed limit in order to be able to do the lazy thing and follow another car.
Having worked on an autonomous car and competed in the Urban Challenge I can attest to the safety of these vehicles. They truly can navigate obstacles that are withing 2-3 inches of the car reliably and confidently. The other interesting thing is that highways are in many ways one of the easiest types of road to drive on. Highway driving while fast, is very structured. There are relatively few rules to follow and they are rarely broken. For this reason I believe highways will be the initial realm of autonomous vehicles and that they will be allowed to drive at higher speeds(perhaps under the guise of human control of the speed in order to shift the blame to a human party.)
Sadly, a lot of speed limits aren't set according to government standard 85th percentile, and local governments are easily swayed to lower speed limits just with a few letters from citizens, no study needed.
That's when you get things like 6 lane divided roads with a 50km/h speed limit. Just excruciatingly slow.
It's important to note that there isn't an obvious relationship between traffic volume and traffic speed. If you model traffic separation as a constant two seconds (front-to-front) your traffic flow is 1800 vehicles per hour at all speeds.
Of course, your average person cares more about how close the back of the car in front of them is, so flow rates at very low speeds are lower than at high speeds. At higher speeds you get other problems, though - traffic gets "patchy" as some people get a little nervous, as people drive at different speeds and so on.
You probably get the most people per hour through a lane of traffic at around 60km/h. Everyone's travelling at about the same separation and about the same speed, and they're all tailgating the guy in front of them trying to make him go a little faster. The speed leaves enough breathing room for everyone to drive consistently, which is supposed to be the big benefit of these self-driving cars.
Of course, these lower limits seem to increase travel times in a very obvious way, and the congestion relief they provide is pretty abstract. It's understandable that people aren't entirely convinced.
The major danger on highways is difference in speed between cars ("variable traffic"), it's much safer to go with the flow of traffic. Absolute speed has little to do with it.
Additionally, speed limits are set for the benefit of police (increasing ticket costs) and for the benefit of politicians who can now say they are "making roads safer".
Well, there are a lot of other reasons for the speed limits as well.
Lack of sufficient climbing lanes on most of the original interstate system. Poorly designed junctions, including merges into the high-speed lanes Inadequate road thickness (autobahn is 55-75cm, US is 28cm), initial quality, drainage capability and maintenance. Inadequate driver training. Lack of enforcement for failing to bear right (though the US is doing this more now) and other minor moving violations that disrupt traffic flow.
The US interstate system would need considerable re-engineering, more funding to keep up with the required maintenance, and both tougher licensing requirements and stricter enforcement of minor moving violations that increase risk. It's a huge cost, that doesn't provide proportionate benefit in all parts of the country, and a questionable ROI. Trust me, I'd love higher speed limits, but I realize the giant mountain of money and effort that would be required to make it happen.
The debate over the exact causes of traffic accidents, fatalities, with regard to speed limits, intoxication, texting, etc. can be simplified by just agreeing that the largest general cause is driver inattention or error, without drilling down any deeper.
Driverless cars are fully attentive, and should have a lower error rate than normal / tired / drunk / distracted / angry / frustrated humans. This I welcome.
Speed limits will, inevitably, climb as safety increases based on self-driving cars and more predictable traffic conditions. Self-driving cars will not just obey the law, they'll also keep a safe distance to the car in front, based on speed and time-to-stop (among correcting many other common mistakes drivers make that lead to more accidents, the faster they drive).
And my state, for one, will have to find another source of revenue. It's politically impossible to raise taxes, or even keep taxes at the same level, in Washington.
I would argue that it is unethical to have revenue tied to police actions. It gives the state incentive to criminalize more and more behaviors, and it encourages selective enforcement (if every speeder were pulled over consistently, there would be no speeding, but that wouldn't be profitable for the state). It also muddies the mission statement for police; "protect and serve...and randomly punish otherwise law-abiding citizens" just doesn't have a nice ring to it.
But, I'm not a fan of the punitive nature of our criminal justice system in general (don't get me started on how many Americans are in prison)...so I may be in the minority in taking this view. And, whether people agree with me or not, actually making the changes in revenue structure required to completely eliminate traffic tickets as a source of revenue for municipalities will be a challenge. Governments are not good at reducing their need for money, so it's likely that most will have to squeeze the money out of some other source, such as higher taxes.
> I would argue that it is unethical to have revenue tied to police actions.
Oh yeah, I agree. But when the voting populace effectively agrees they'd rather take a chance on not being pulled over for speeding rather than approve an income tax for people making over $250,000 a year...well, you reap what you sow.
I suppose that new use taxes will be created for driverless cars. Especially driverless and eventually passenger-free commercial vehicles. The state can take a share from the resulting productivity increase.
As a thought experiment, imagine that Walmart is given the opportunity to buy more expensive driverless long haul trucks, that are still much cheaper to operate because no driver is needed. Walmart would be willing to pay a use tax for each mile driven as long as the total business case made sense for the life of the new vehicles.
I hope new taxes don't happen. That would set back the move to driverless cars...every year we don't make the switch is a year in which tens of thousands of people die in car accidents that didn't have to happen. I suspect this is one of those areas where we're only thinking of the negative consequences, and not the positives.
What if you no longer need highway patrol officers to enforce speed limits and traffic laws? Or maybe you need a tenth as many on the road?
What if the number of times emergency responders are needed dropped precipitously because of fewer accidents? Emergency room visits are the only socialized medicine in the US (and horribly inefficient socialized medicine at that), so reducing them will have a huge impact on road-related expenditures borne by all of us.
What if we can begin to phase out signage and signals because the cars know where we're going and know when to stop and start? It will be decades before all cars are self-driving, but roads and lanes could be designated self-driving only, and those lanes could be free of extraneous signals and signs.
What if we can reduce number of miles traveled by optimizing routes, thus reducing damage to roads (subtly, but for big trucks, the differences could add up faster)?
Taxing self-driving vehicles more would be an impediment to a great good, so I hope that doesn't happen.
Once (begin assumption here) autonomous cars become the norm I don't think this will be a problem. The reason speed limits exist at all is because we don't want people who are not physically capable of driving at that speed (poor eyesight, bad reflexes, etc....) to try, so we set the limit at a level where we think everyone should be reasonably ok in that situation.
But if all cars are run by computers, we can raise speed limits considerably, since reaction time, sleeplessness, stupidity, etc... are all non-factors anymore. The only problem is whether or not your car can actually go as fast as everyone else, but for that we can set up fast and slow lanes similar to the somewhat unofficial way things go on the highway now (faster the closer to the leftmost lane you are).
> The reason speed limits exist at all is because we don't want people who are not physically capable of driving at that speed to try
Is that the reason we have speed limits?
I've always thought it's more like we don't want to deal with the consequences of something going wrong when there is too much energy involved. i.e. blow a tire, or hit a patch of gravel at 60mph and see what happens. Now try it at 160mph.
I'll be fine with that as long as lane splitting is legal on my non-autonomous motorcycle. Since lane splitting is legal in CA, I wonder how well google's vehicles deal with other drivers that split lanes?
I've seen demonstrations where the autonomous car recognizes bicyclists and adjusts the travel path to avoid the "obstacle". I imagine with a motorcycle the autonomous car should always see it coming up behind it and could be programmed to give it some room as it gets close.
This will be an issue as long as we have more human drivers and few self-driving cars. In future when there are more or all of them are self-driving ones, there won't be any need for limits - speed or otherwise. The cars can know the position, speed of each other and figure out how fast they can go to reach their destinations. Perhaps you can pay a subscription so you can always drive at a higher speed than most normal people or buy tickets to drive fast for that day when you need most.
Remember that most of the world hasnt even adopted automatic transmission because people like to have full control over their cars and that technology is decades old.
Once we get to the point where all cars on highways are automation capable, if not outright enabled, there should be the option to have speed dictated by road conditions so that, under optimal conditions, the system could adjust speed up, and adjust speed down as conditions worsen. That is, allow speed to be variable and dependent on conditions on the road (at that particular section). Or, they could rate different lanes at different speeds, or a combo allowing people to tell the system where to merge into.
There is the complication of less capable cars --so that maybe bunches of cars (those part of a wave of traffic) would have to submit to the safety levels of the least safe vehicle. (Let's say a bunch of high performance sports cars and an economical car which isn't designed for hugging the road nor has grippy tires; or, throw in a cargo truck).
Maybe the rightmost lane could be reserved for non autonomous vehicles like classics and holdouts (people afraid of automation/driverlessness).
Given that speeding tickets would be rare, the state could impose fees based on a combination of emissions and also road wear (tonnage, for example) to offset lost revenue.
Part of the problem is that communities keep lowering the speed limits below recommendations and even the guidelines of law. My father, who has a background in urban planning, is fond of, or frustrated enough to point this out.
A street that, per our state's code and regulation, should be 45 mph, gets dropped to 40, then to 35. Nothing significant has changed with its facing development and use. But the community, or of some excess of caution or some orthogonal desire for control, keeps ratcheting it down.
In other spots, they start putting up stop signs at every block. They aren't really warranted, but in a variety of "think of the children" attitude, perhaps combined with a twist on NIMBY (Not In My Back-Yard) attitude, the through-street needs to "be somewhere else". Until there no longer is a through-street.
So fine, have these cars obey the speed limits. But then set reasonable limits, ones that are evidence-based and not politicized.
Road speed limits are based on standards that take in to account the environmental circumstances of the road - straightness, visibility, etc - as well as having a huge amount of safety buffer built in to account for human reaction time.
If self driving cars become ubiquitous, there is no longer a reason to have this buffer built in. Perhaps the whole idea of posted speed limits will become obsolete if the software determines the maximum speed in the moment for a certain risk tolerance.
One thing that always worried me about self-driving cars - what about an emergency? For example, hypothetically, if my wife was giving birth and I needed to rush to the hospital, would I have to poke along at 50km/h and hope to get there on time? There are times when it's perfectly justified to break the speed limit (other examples: outrunning natural disasters, going to visit a dying relative, and so on).
You call an ambulance, that has an overclocked version of the self-driving chip :).
I don't see the self-driving car as a completely different car than a normal one - maybe more like an expensive option (with all the sensors, data and stuff it will be expensive) for a normal car, that can give you back the control when you want it.
No it is not perfectly justified to break the speed limit if you are going to visit a dying relative. That's ridiculous. Think about it. If I was going to visit a dying relative (who you have never met) would you be happy for me to drive at 100mph along the street where you live?
A self-driving car would be handy in situations where you are alone and injured, with the capacity to get to your car but not to drive safely. Severe illness, power tool accident, bear attack on a camping trip. You just get in your car and say, "Take me to the emergency room." It would actually be faster than an ambulance that had to make a round trip.
Also, pretty much the last person we (as a society) want speeding is a tired, stressed, freaking-out, distracted, soon-to-be-husband.
It's a disappointment this writer is covering this topic.
Self-driving cars can be great because of the shear impact on utility for vehicles and safety implications.
I'm quite excited about the advent of self-driving cars and the underling operating systems that will make them amazing for humans.
I do believe there is room for both types of driving. It's possible our freeways will be segmented such that automated cars can go in certain lanes (perhaps even barrier protected) while the remaining traffic stays in other lanes / roads.
Traffic, or more specifically congestion, is primarily caused by excess cars on any edge (road) in which it simply takes N+1 cars to cause. Inconsistent driving patterns cause effective waves of inconsistency down the pipe exacerbating the situation (stop and go traffic actually makes congestion even worse).
Thinking about the overall network at a higher level makes all of this even more exciting. Consider the scenario in which N cars are on an edge with capacity N. If your car was about to enter the road you would basically contribute to traffic. This cost, C, is then felt across N+1 (don't forget to include yourself!) people at a minimum which is C*(N+1) minutes of human time that is lost.
If our cars, and ourselves, were better informed you might take an alternate and sub-optimal route that taxes you an extra T minutes but the overall network of time is optimized. Perhaps you'll even get a reward for it. And perhaps the reward is simply allowing all the people currently on the road a faster commute home so they can be with their loved ones.
Beyond that, as people have mentioned here already, the faster we can go and caravanning (which can be done safely by allowing cars to attach to one another in train format) improves overall throughput (and make traffic less likely).
Finally if we think about traffic, it's really just load and it needs to be distributed better across all possible routes and time. We have commute hours, but one thing to think about is that with a self-driving car I can leave to LA at 12AM and get there at 6AM while sleeping in the back of the car.
Self-driving cars will increase utility on vehicles, but also on time for travel routes ideally. This will spread the load (since humans generally have some biological clock they operate by) to utilize the 24 hours a day we have a little bit better.
And if your car can't speed I guess we should invest in really good alarm clocks and reminders :)
Not when you can't yet afford an autonomous car and are cursing at the guy cruising along at 50 ahead of you, getting work done comfortably while you're gripping the wheel in frustration...
If self-driving cars keep their much safer track record when they start being actually used by the public, maybe they can adapt the carpool lane to be a self-driving car lane which has a higher speed limit?
I think the first usage of self-driving cars is for public transportation.
With this technologies, crowed areas could have more self-driving buses.
or use it to replace long distance truck drivers.
Why should the robot care about the honking? It should already have way more data via sensor input than the human doing the honking has long before the honk ever occurs.
Even ignoring that, I don't think I've ever seen a situation in which I think a horn actually helped in a near-accident situation. It tends to just be the thing someone honks in anger after the immediate danger is already avoided. IME (as someone who has never been in an accident but has seen quite a few near misses and accidents on SoCal freeways) usually once the horn comes into play it is either too late and the accident occurs anyway or the horn is honked after the fact as a "fuck you" to the other person.
Actually it should care about signals. Maybe not honking, but there are cases like emergency vehicles that you react to. I'm not sure what's the standard reaction in other parts, but in the UK when you hear / see an ambulance or a police car behind you, you stop on the side of the road. Sometimes partly on the pedestrian area. Would the self driving car do the same to let the emergency vehicle pass it? Unfortunately that's unlikely.
This will no longer be a problem if all cars are self-driving. Then, they can just up the limit incrementally. Possibly even to something much faster than a human could do manually.
The idea that we allow almost anyone 16 or older to control several thousand pounds of metal that can travel at high speeds really is quite absurd when you think about it. There are more than 30,000 deaths every year due to car accidents, and hundreds of thousands more injuries.
It's amazing how screwed up our society's priorities are. That's ten 9/11's worth of deaths every year from car accidents. By the time the wars in Afghanistan and Iraq have ended, we'll have spent around $4 trillion on them altogether[0]. If even 1/100th of that had been redirected to research on self-driving car technology, we might have already saved tens of thousands of lives.
0: http://www.reuters.com/article/2011/06/29/us-usa-war-idUSTRE...