> In late September, a Waymo spokeswoman told Ars by email that the Phoenix service would be fully driverless and open to members of the public—claims I reported in this article.
> We now know that Waymo One won't be fully driverless; there will be a driver in the driver's seat.
This reminds me of this famous Radio Yerevan joke:
> Q: Is it true that Ivan Ivanovich Ivanov from Moscow won a car in a lottery?
> A: In principle yes, but:
> it wasn't Ivan Ivanovich Ivanov but Aleksander Aleksandrovich Aleksandrov;
> he is not from Moscow but from Odessa;
> it was not a car but a bicycle;
> he didn't win it, but it was stolen from him.
Because smart people have invested billions of dollars into this technology and because even smarter people work on this technology the general public must pretend that it's all going according to plan, that everything is damn serious and that the emperor still has his clothes on.
The chairman of a local chamber of commerce had to introduce the speaker at the organization's annual black-tie dinner...
"The man who I am about to introduce," he said, "is someone I know you'll enjoy listening to. He is the most gifted businessman in the country. He has made one hundred million dollars in California oil."
The speaker, embarrassed, came to the podium. "Thank you for your kind introduction, Mr. Chairman," he said.
"However, the facts need some clarification. It wasn't oil; it was coal. It wasn't California; it was Pennsylvania. It wasn't one hundred million; it was one hundred thousand. It wasn't me; it was my brother. And he didn't make it; he lost it."
To me it sounded like they basically just gave people the option of getting out of the NDA if they wanted to start paying for rides. I assume to most of the early riders, it's just not worth it when they can continue just getting free rides.
I doubt that it's human drivers that are the issue. Cars are nice big easy to identify predictable things, and even if you screw up and hit one the cost is relatively low.
The challenging cases are probably things like children playing on the side of the road, pedestrians, bicyclists, construction workers, cops standing in the street, people working in manholes, and so on. Relatively small, easily obscured, unpredictable, hard for computers to figure out how they are moving, and very high cost to hit.
No. The problem is that road systems (even really easy ones like in America) are designed for humans, and humans are requires some human abilities that AI isn't even close to, like anticipating what people are going to do.
Also I think the actual hardware is not good enough yet. LIDAR is great but humans can see really far in really good detail and can work out what is happening even in really bad conditions (rain, looking into the sun, etc).
Non-motorway driving is just too varied. I still think the first real application of autonomous vehicles will be driverless lorries that drive between depots that are connected by motorways. It's way way easier because the environment is so restricted, there's a clear commercial need, and people would be a lot more comfortable about the safety aspect - who hasn't seen a lorry drift into the hard shoulder? Plus lorries are super expensive so it wouldn't increase the cost too much, and they're huge so there is plenty of space for hardware.
This is a series of good points. Near where I live (Australia), we had some roadworks on the freeway, causing the lanes to be temporarily moved. The "new" lines where yellow, instead of the normal white, and much harder to see. Further, this happened several times, so you just followed the "most yellow" line, and hoped that nobody rammed into the side of you.
If, instead, we have a set of local sensors setup that identified the new "safe" path, an automated car could easily go through this new section. As marked, however, it would be very dangerous to rely on an automated driver.
No. It's that humans are remarkably adaptive and can convey intention through body language -- even through a "puppet body" like a motor vehicle. Humans can, by and large, adapt to "terrible" drivers -- swerving to avoid the aggro asshats, honking at phone addicts to signalboost green lights, changing lanes to avoid grandpa's leaden brake foot, etc.
Speaking as somebody who owns a classic car, drives carefully to baby said classic and often rides shotgun... I gotta say: most drivers are pretty decent. The people I hear complain about how awful everybody else is... they're either old and drive dangerously slow, or they're super entitled and drive aggressively. Since we're on HN, I'm gonna guess you're of the latter camp. Do the world a favor: take Uber and don't write AI.
I don't get paid by-the-hour to drive; I get paid to arrive on-time. So commuting in my opinion (like IT in MBAs' opinion) is purely a (time-)cost center. And from that POV the average driver is indeed absolutely terrible. Ever since I started working in the software industry, I don't think I've ever taken the wheel of my car and driven it somewhere, without getting angry at another driver along the way.
Granted, different drivers can be terrible in different ways.
Some of them do stupid mutually-harmful BS like pulling out right front of me with minimal margin and no intent to travel at even half the speed-limit.
Others think that because they have an 8-cylinder engine and a traditional automatic transmission, that they should rule the streets - which they can and do, but that doesn't make me any less angry at them when they e.g. pass me at a 20mph relative-speed on the same side into which I was signalling to change because I was behind a schoolbus.
I'm not offended, I'm bemused that you couldn't resist attacking me while making your point.
And no, talking about the avarage human being's ability to operate a specific machine over the course of their life is not painting any individual with a broad brush.
Also, an individual can be a good driver. People in general may not be. That's not a contradiction or an insult. Heck even a "good" driver can drive badly.
> The people I hear complain about how awful everybody else is... they're […], or they're super entitled and drive aggressively.
No, I simply expect others to follow the same laws that I'm bound to and that I follow. The Golden Rule. And a lot of drivers on the road simply do not do that.
> humans are remarkably adaptive and can convey intention through body language -- even through a "puppet body" like a motor vehicle. Humans can, by and large, adapt to "terrible" drivers -- swerving to avoid the aggro asshats
I agree with you here, but this doesn't really help demonstrate that humans are not, on average, poor drivers. That some of us can compensate for the shortcomings of others is great, but the fact that we not only compensate for the shortcomings, but that we're able to suss it out of such fine data — seems only to be evidence that we have a need for it, because collectively our driving is bad. Perhaps you can teach an AI to read the "body language" of a car, but I imagine that such a thing would be remarkably subtle and considerably difficult.
(On the whole, I don't think it really matters what the skill level of humans are. The AI must perform better, or else what's the point? Ideally, we should be able to prove that it's better — so as to sway public opinion, and I suspect that will be the hard part.)
> "Lane changes appear to be a problem for the cars," Randazzo says in the video. When trying to move into a crowded lane, a Waymo car seemed to lack a human driver's ability to anticipate other drivers' actions and squeeze into an open spot. Instead, the vehicle would turn on its turn signal and wait for a few seconds for an opening to appear. If one didn't appear, it would turn the turn signal off and wait for a while before trying again.
This makes sense, because a lot of times merging into a lane is a game of chicken. You've got to make the drivers in that lane believe that you're going to merge anyway and that they'll need to slow down to avoid hitting you.
Always curious how traffic is viewed as an antagonistic activity by many. I usually have the opposite impression. Someone wants to switch lanes in dense traffic, so I slow down slightly and let them slip into my lane. Others do the same for me. That's working with each other not against each other for me.
I had an accelerated course in aggressive driving over a couple of hours when going through it. In comparison driving in SF or NY is like visiting a spa.
Sorry, but no. LA is really that not that different from NYC, northern NJ, DC, Boston, Chicago, SF, etc. Traffic is bad in all those cities and drivers are all pretty aggressive. I’ve driven in all of them and never noticed any real difference in the difficulty.
Now, driving in southeast Asia or southern Italy was very, very different from any US city.
That sounds so polite. And unlike anything I have ever experienced. Especially in the truly dense traffic, cooperation means giving up just enough space to let the other guy take it, as long as he's assertive. Definitely not a place for the timid, I've found that people who are accustomed to driving in extremely dense traffic tend not to have much patience for lollygagging. It screws up the rhythm.
I leave massive gaps constantly now, it lets me see farther ahead so I can switch lanes when I see a problem coming and my average trip time is definitely shorter because of this.
I really like this video for describing some of the advantages of "zen driving":
in LA, drivers on a side street who are waiting to make a right hand turn to get in your lane don't actually wait. they just start turning and enter your lane.
you better slow down and give them an opening or you'll hit them. which you will definitely feel like doing sometimes.
that's one reason i really look forward to the day when cautious, defensive, courteous self-driven cars replace all human drivers here.
Well thing is that if you put on the blinker, I’ll start leaving a gap. But if you don’t take it I’m gonna stop leaving one and assume your blinker was a mistake.
... but as soon as I can recognize its Waymo autonomous car, I stop caring and continue driving, because I can easily assume how machine will react (full stop, etc); meanwhile driving on the road around human drivers I have no other choice that assume most or majority are unpredictable; so I better stop let them in, before some crazy loon do something we both going to regret.
Honestly looks to me like the AI translation problem, always 5 years away from being perfect, still mostly crappy, only good enough to have a rough idea of what is being said and with glaring mistakes.
AI translation has gotten massively better in the last 10 years. I can read a foreign news site in Chrome, pop "translate this page", and get a reasonable sense of what the article is about and the factual content it contains. The grammar will be off and some sentences will be non-sequitors, but getting to this point is a massive leap forward compared to 15 years ago.
I feel like a lot of AI problems move directly from “this is so terrible that it’s completely useless” to “this capability is a fact of life that I no longer think about” without ever passing through a conscious “this doesn’t suck so much anymore” stage. Speech recognition and translation both did this. They’re far from perfect but they’re good enough to get a lot of real-world use, but somehow they still have a reputation for being worthless junk.
for some people, waymo-like cars are not an optional luxury/curiosity but a necessity (elderly people, disabled). It would be nice if they focused on those groups early, instead of leaving it as an afterthought. The oddities of the cars are not going to be a show-stopper, as long as the cars are safe. Human drivers will learn to adapt their way around the "roomba cars" over time.
it s going to be a huge market if it proves doable. Consider for example wheelchairs that pair with the car and auto-dock (or even self-driving wheelchairs). selfdriving cars are robots that just happen to have wheels, the posibilities are endless
As a taxi service, it indeed won't be much different. But once the technology is well established, I presume there would be options for leasing. Having the option to just go wherever I want, even longer distances, would be amazing.
It's strange to claim that a product that doesn't yet exist is a "necessity". A necessity is a good critical for survival, not just anything that grants a quality of life improvement.
Hadn't really considered that. Where I live, the transit company runs a van service door-to-door for elderly and people with disabilities that cannot reach a normal bus stop. The cost of the tickets are the same as for the normal city bus.
What do disabled folks do now in places that don't have such a service?
They did focus on these early though. Literally their first fully autonomous drive on public roads was with a blind individual, explaining how the technology was going to be transformative.
They've been working with the guy in the video for over 6 years [1], and he's the former CEO of the Santa Clara Valley Blind Center, which has aims for "a future where awareness, legislation and technology meet at a crossroads to provide individuals who are blind and visually impaired the same opportunities as their sighted peers" [2].
Autonomous cars are a very difficult problem, more than I think most people anticipate. There's been a lot of tax breaks/money that goes into these programs too, and that's the tragic thing, since that funding should be going to known solutions: public transportation (trains, trams and express buses).
Even if you had a full-autonomous only Interstate where each car could be fully loaded and ride bumper-bumper at over 100kph, it wouldn't even begin reach the carrying capacity of currently existing fully autonomous subways systems like those in Singapore.
These driverless cars would be great in Europe where they can solve the last leg problems in countries that already have strong public transport infrastructure. In places like America, the existing infrastructure has a long way to go before autonomous cars won't lead to the same gridlock as normal cars .. and an even longer way to go before they're ready for mass consumption.
> There's been a lot of tax breaks/money that goes into these programs too, and that's the tragic thing, since that funding should be going to known solutions: public transportation (trains, trams and express buses).
Citation needed.
> Even if you had a full-autonomous only Interstate where each car could be fully loaded and ride bumper-bumper at over 100kph, it wouldn't even begin reach the carrying capacity of currently existing fully autonomous subways systems like those in Singapore.
A train or subway car can carry at least 10 people per meter vehicle length, while a car is usually some four to five meters long and usually carries only one or two persons. Of course, the comfort factors are decidedly different.
I've yet to hear of a subway where there is no space between consecutive trains, however. If there's only one meter of train per hundred meters of track...
With true self driving technology it's plausible the cars on the road will be notably different than what we have today, which might significantly increase the roadway carrying capacity.
That's true, but it doesn't resolve the disagreement. Consider the fact that trains aren't infinitely long, and so they have to slow down everyone to let some passengers off. There's an idea where a train "car" is boarded and accelerated separately then joins the rest of the train while it's moving, but this approach is rarely implemented for many reasons.
> and so they have to slow down everyone to let some passengers off.
So do roads, slow downs happen when cars get on or off of highways.
Going from a road with a high speed to one with a lower speed, if those roads are at capacity, will always result in a backup. Off ramps serve this purpose of course, they basically convert cars going at 60mph into traffic going 30mph through stop lights or some other sort of flow control, but if the roadway is at capacity (as many major cities are), that off ramp as a flow control mechanism falls apart and cars pile up onto the freeway.
The math is unfortunately independent of self-driving cars. If every car is self driving capacity does increase, cars can self coordinate and some of those stop lights become unnecessary, but if 100 cars want to get onto a road that has room for 20, someone is going to have to wait!
This argument seems to compare 1 interstate hwy to 1 rail. In fairness to the autonomous vehicles, there's many road routes to most places. Road infrastructure costs a tiny fraction of rail infrastructure.
Roads also allow for multiple lanes. This allows for somewhat degraded service when something goes wrong; a train stopped on the tracks blocks everything. With multiple lanes, roads can be maintained while still in service. Every subway struggles with maintenance windows.
I'm not saying one is always superior to the other - trains have their niche, but it's a niche that will never be as broad that of autonomous cars.
We definitely need at least four train tracks for mass transit to work. Two tracks each way: one express and one local.
What I don't understand is why do autonomous self driving cars and subways have to be either or. Why can't we do both? After all, we can't use the subway for ambulances and fire trucks so we still need to build some kind of roads even if we had full confidence in the subways...
Just thinking out loud... Unless we had helicopters for emergencies?
Can we have a town with no paved road? Only sidewalks? That's still paved, right? Can trash collection also take the subway?
> Road infrastructure costs a tiny fraction of rail infrastructure.
Not necessarily.
Rail is cheaper to maintain, and there is a huge opportunity cost of paving over a cities most valuable real estate to make roads.
The math on how large the streets would need to be in downtown Seattle to alleviate traffic end up with a giant % of downtown Seattle being torn down and paved over.
Rail can also have multiple tracks, it costs more, but amortizes really well.
Not having to intermix pedestrian, bicycle, and cars is another huge win for, well constructed, rail projects. Autonomous cars still have to stop at crosswalks! In urban cores, stop lights seriously cut into average MPH for a trip, and start shifting things into favor of rail.
He said an interstate fully loaded. A fully loaded interstate has 10 lanes (5 on each side), and with 10x cars, each 3 meters long with 5 passengers, that's 16.6 people per meter.
Moving 100kph continuously bumper to bumper, not having to stop to let passengers on or off, my bet is that the interstate has more throughput than a subway train.
> Please don't post shallow dismissals, especially of other people's work. A good critical comment teaches us something.
Handling 30k passengers per hour is manageable for most rail lines around the world.
Torontos Bloor line does 500,000 daily.
Similarly highways are constrained to their own maximums per lane, and counterintuitively raising the speed limit doesn't help. 1500-1700 vehicles per hour is a fair estimate.
He posted unsubstantiated claims about tax breaks (I'd like to learn where there are tax breaks for autonomous vehicles) and a claim about trains having more throughput than an interstate with cars fully loaded moving 100kph bumper to bumper.
Asking for citations might be a shallow dismissal, but stating hypotheses without facts to substantiate them is also shallow.
That's true, but taking the thread further in the wrong direction only makes it worse.
There are other ways to ask for more information, that open discussion up rather than stiffening into confrontation. Please post in that spirit instead.
Everywhere and everyone is different, obviously, but I would love to see autonomous electric vehicles become our future. I really don't care for public transportation, I find it uncomfortable in many ways, and I don't see any way that it will ever not be about half the speed of a private point-to-point transport.
So a highway filled with self driving cars zipping along bumper to bumper may not technically be able to carry as much as a fully loaded subway train, but it's so much more useful. Take away the casualties and pollution, and it's as close to perfect as we are likely to get.
The lack of comfort people refer to usually involves the other passengers, not the amenities. But everyone knows you're not supposed to talk specifically about that aspect of public transportation.
Ha, I will! I once had a guy behind me on the train whip out his penis and pee on the floor. Looked right at me in the reflection of the window as he did it. He wasn't even crazy, either, as far as I could tell, just antisocial.
And yes, before you say it, that isn't an everyday occurrence. But jackasses refusing to take their bag off the seat so you can sit down is. As are people who can't be bothered to take a shower, or think that cologne will make up for it. The list goes on. Many people are just unpleasant to be around, and I don't see any problem with that observation.
For me, public transit comfort comes down to a couple things. First, the hard plastic seats make my butt hurt, and standing makes my left foot ache. Sorry, I'm getting old :). Second, the 2x time it takes to get anywhere means I'm on that terrible seat for even longer. And third, because yes, it's absolutely true, I find that perhaps one in ten people is unpleasant enough that I want to be out of visual and/or aromatic range.
> First, the hard plastic seats make my butt hurt,
Yeah, same for me. And I've heard that the ~80s crop of NYC subway cars were made in Japan. Except there, they have upholstered seats. But that wouldn't work out well in NYC, so they got hard plastic. But on the other hand, the oldest LIRR cars (refurbished for the 1966 World's Fair) had upholstered seats, and they were ~OK.
Yes,precisely. I found Vienna and Tokyo’s subway pretty immaculate. But London, Paris, Bilbao, Amsterdam all had some experiences similar to my other post. Much smaller sample so hard to say how frequently it occurs in any useful way.
I still think public transport is a net positive. But completely understand the allure of my own car. Or an Uber managed self driver where each rider rates the past.
My car has WiFi, and it is not even automated yet. And food & drink are perfectly feasible as well, remember we're talking about automated cars now. All kinds of options suddenly make sense when you don't have to worry about the chore of driving.
Some Finnish trains have play-areas for children on them. I was pleasantly surprised the last time I rode a train with my 1.5 year old child. He had a lot of fun going up and down the slide, sitting in the toy-train (on a train!) and reading books in the "library".
Except in lucky places with great underground railways you basically have awkward last-mile journeys where you have to take a taxi or hit a car or take a bus and that adds a lot of time to the journey.
It would be great if there were trains everywhere but frankly that's even harder than making a reliable driverless car.
Which is why driverless cars are great for bringing people from and to public transport. They are a great ingredient for improving public transport, but they don't replace it (at least as soon as road capacity becomes an issue).
How do I get from my house to the train? Do I have to make transfers? How long will they take? What if I have to carry groceries, or take young children with me, or my elderly grandma?
If I have a car, I load everyone and everything up, and we go. We get to our destination and unload, ta-da! Is this not an order of magnitude more useful than taking a train?
Sometimes I think HN is filled with young people with uncomplicated lives :).
>Sometimes I think HN is filled with young people with uncomplicated lives :).
No, if these are big issues for you, you likely live in an American low density area where public transportation is scarce. In large metropolitan areas (which is after all, where most of the world already lives, and will live), getting on and off public transport is usually not a problem.
I have relocated to Tokyo for example about a year ago, and for me, a trip to the grocery store is usually done by foot (same is true for other appliances), and multiple forms of public transport are in walking distance.
I live in a city with about 700,000 residents inside the official limits, and 3,000,000 residents in the metro. The nearest bus stop is approximately a quarter mile away.
With all due respect, the number of people who live in cities as dense as Tokyo are the minority, most of us don't.
No offense but if you think walking a quarter mile is a hindrance to access public transport then the car culture has done even more damage than I thought.
On the contrary, I think a quarter mile is perfectly reasonable for a public transit stop in my city. Sorry to suggest otherwise.
However, when I think about loading up my young children with all of their stuff, loading me up with everything I need, etc, and then dragging that down the street a quarter mile, then waiting for the bus, boarding it, riding to the transit center, transferring, riding another bus to the light rail station, riding that downtown, and then doing it all again the other direction... well, I'm pretty confident in my original point that the car is a hell of a lot more useful. Sure, you can exist with only the public transport option, but I was not asserting otherwise. But a lot of people on HN seem to think that not only is it possible, but that it is the superior choice.
So, no offense taken, in fact I appreciate you affirming my original comment about so many folks on HN commenting from the viewpoint of someone with a young, uncomplicated life :-).
I think you're making it more complicated than it is in many places. In many you'd need not get in a bus at all, because the rail stations are dense enough to always be a couple blocks away. That requires having more than just 1 train line though, which is only really true in NYC (and maybe Chicago? You have to squint really hard for but to be the case even in sf) if you're looking in the us.
Many people with kids do what you describe in cities. It's far better for everyone, and even if it's not the superior choice for an individual (though I'd contend it is when you adjust your lifestyle), it is the superior choice for everyone together as it avoids pollution, traffic gridlock and huge fuel inefficiencies.
As it is those with cars push off these negative externalities onto those who choose not to have them, or can't afford them. So your comfort does have a cost, which you don't pay directly.
Perhaps I'm sheltered or naive, but it seems to me that in a well designed system you shouldn't really have to take two bus connections to get to your nearest light rail station. More like zero, or one max if you live on the outskirts. This is how it is in London, and while I realize that's not a typically sized city, it's not unusually dense either. I think that kind of system is possible anywhere, it just requires motivation.
¼ mi is a 5min walk. Often you need to walk as far or further to and from the parked car. I literally yesterday just took 2 kids, a toddler and a baby, on a multi connection trip, and it's easy, simple, and less stressful then driving.
Also, tour last paragraph just makes you sound pompous and liek you have no intention of having a discussion but instead only wanting to put others with a different point of view down.
A quarter mile to the closest route typically means a far more complex combination of transfers, each with its own delays, to actually get to your destination of choice.
A quarter of a mile is a long distance for a bus stop in a city with good transport. But sounds like your city doesn’t have enough transport density. Tokyo is pretty dense but its still not abnormal.
I doubt that a train is significantly more energy efficient than an electric car with 3-4 people in it. Most people will probably use pool type autonomous cars which get you to your destination door to door much faster than trains.
> I would love to see autonomous electric vehicles become our future. I really don't care for public transportation, I find it uncomfortable in many ways
I imagine (edit: many) people who use public transportation would agree with you. You now, since they're usually using it out of necessity and not enjoyment.
Is that true for anything less than a subway/metro train? Light rail is slow, buses even worse, there are many benefits to public transit but speed is rarely one of them. And in the rare cases when it is faster, it's in a handful of extremely dense cities that only represent a fraction of the population, and only for specific use cases.
Right. We get that. It's very obvious that much of the hype around self-driving cars is premised around people who simply reject mass transit outright and therefore believe that somehow, somewhere, they'll get to ride around in a spacious, individual vehicle in the middle of a packed city and so will everyone else and somehow it will all work out because "technology".
I admit that I'm only doing rough math in my head, but it seems to me that cars bumper to bumper at 100kpm is definitely going to have a greater carrying capacity than intermittent larger vehicles that stop a lot. (Or at least a similar magnitude)
I don't expect self-driving cars to drive much more densely than current cars, though, because all or almost all the same reasons we drive with separation still apply for self-driving cars. Where you might think you get some advantages, they will also have disadvantages, too, like being able to compute exactly how risky their behaviors are. People making the cars and writing the car software aren't going to want to stand in court and explain why their car crashed the 50-car dense train because their car software left no possibility of the brakes going out at that exact moment (i.e., no previous sign of brake failure), and caused a huge crash because none of the cars had any buffer for failure.
Personally I expect self-driving cars to drive a great deal more conservatively than humans as a result, and I think self-driving cars are going to end up demanding a lot more space and buffer on the freeways. And if they don't at first, one day you'll way up and they suddenly all will, because somebody won some lawsuit and a software update was pushed overnight to double the buffers.
I tend to agree, I expect that as a practical matter autonomous cars will be conservatively driven, e.g. with large buffer zones, gentle acceleration, safety-above-all-else.
I think other drivers will adapt to this and start taking advantage. Think of what happens when you leave a legal following distance in front of you now, in traffic. Now imagine an autonomous car in the same situation. It would probably end up stopped.
I'm not sure this is a bad thing. Often times, traffic jams are caused by not enough spacing, or unobservant drivers resulting in sudden breaking, and a knock-on effect. Would need to simulate to confirm, but I suspect that more vehicles giving enough stopping distance, and accelerating smoothly would improve traffic flow, even with the addition of more defector (defective?) drivers.
The fundamental difference between human and automated drivers is that humans aren't going to get any better. Software based systems will get to human level then continue to improve to the limits of computing and physics, which is certainly better than humans.
And we haven't even considered networking. When the cars are all talking to each and routed together then it's fundamentally a single machine.
The laws and systems will evolve with capabilities rather than getting stuck on supporting the earliest generation of self driving vehicles and norms.
The major reason we drive with the space between cars that we do is human reaction time. That is at least an order of magnitude better in autonomous cars.
The primary reason we drive with separation is our reaction time. Computers can react must faster. In addition they can actually talk to the cars around them. They could literally form a train.
We won't switch to self driving cars until they can drive better than humans. If they have to leave a larger buffer, then the work isn't done.
There's still a chance an automated car encounters something that couldn't be spotted / reacted to. For example a deer/kangaroo/whatever jumping right in front of it. If the cars keep their distance, only the first one crashes. If they form a minimum-spacing train, the only message they can send to the following ones is "you're screwed now".
A deer would be unlikely to jump into a train of cars. They usually get caught by cars that have space in front of them. I know you are just giving a single example, but anything that applies to an actual train would apply to a virtual train as well. I don't understand why not having a physical latch between cars makes any real difference.
A deer doesn't affect a real train that much due to the mass difference. In case of cars though they cause serious damage. If the following cars don't keep a distance they end up immediately crashing into the first one, multiplying the damage.
I don't have much experience with deer behaviour, but kangaroos will happily jump right into anything moving. They will even hit the side of your car sometimes (which they should've obviously seen coming)
> and I don't see any way that it will ever not be about half the speed of a private point-to-point transport.
Subways / trains in Tokyo are faster than driving, same in London.
No stop lights, no intersections, no delays on the road.
Buses aren't faster, but once you go underground, the speed improves dramatically. Even if there is a 2 minute stop once every mile, that is still less time waiting than comparable stop lights in a car.
As far as I can tell, that's only true in fairly unique situations, though, in the super-cities. I live in a decently sized city (metro population of perhaps 3 million people, around 25th in the US I believe) with a transit system mostly regarded as 'pretty good' (but no subway, just light rail and buses) and it would be rare even at rush hour for a car not to be faster. Usually signficantly so.
I think what it comes down to is that if you have a subway system, that will probably beat cars, in the limited area it serves. Once you have to do transfers to buses, it gets much slower. In the places I've visited with subway systems (SF, DC primarily) this is definitely the case.
US transit is heavily optimized for cars, even in places with pretty good mass transit. Don't let your personal experience with US metros color your expectations for what public transit can be.
> with a transit system mostly regarded as 'pretty good'
That's probably US bias, because as far as I can tell by reporting US public transport is generally very bad and far worse in big cities compared to much smaller cities elsewhere.
Anecdotally I commute to work because it's faster and much less stressful. Granted I do need to cover the first leg by car due to the public transport in my area, but the little bit of walking and the free time I get in the train are so much better than sitting alone in city traffic.
> I don't see any way that it will ever not be about half the speed of a private point-to-point transport.
Private point-to-point loses out mostly for long-distance already, short-distance currently either allows last-mile, station overhead, and related concerns to big down shared transportation or suffers from sharing infrastructure optimized for private point-to-point.
Ubiquitous automated (on-demand, not personal cars actually make the station problem easier, since you need less parking at stations, making station footprints smaller, and reducing delay from arrival at station to boarding.
They could also merge with buses: instead of fixed routes and schedules, vehicle sizes and routing are demand driven for efficiency (perhaps allowing you to pay a premium to guarantee a dedicated vehicle for your party.)
I would be comfortable saying that subways and above (e.g. high speed rail, airplanes) are a special case of public transport that will always make sense and will nearly always beat out a car for transit time.
Aside from that, I agree with what appears to be your premise -- that we could invent public transport devices that mitigate and perhaps mostly erase the advantage of the car.
But then I'd ask ... why? Why are we supposed to hate cars? It's okay to like scooters and bikes, but nothing larger until we get to a bus? I imagine a future with all sorts of automated cars that never get into an accident and roll silently along emitting not a whiff of noxious fumes, while being big enough to carry just the right amount. If nobody needs to own their own car, then they can call a car that is sized for their exact needs. One passenger or ten. They efficiently get to exactly where they are going in the minimum time possible with minimum effort. What a dream!
> I agree with what appears to be your premise -- that we could invent public transport devices that mitigate and perhaps mostly erase the advantage of the car.
That's not quite my premise.
Rather, it is that shared self-driving automobiles (cars through vans through buses) with centralized automated dispatch are the missing element of the mix of public transport devices that, alongside existing options, does that.
> Why are we supposed to hate cars?
Recognizing that public transport has social benefits does not require hating privately owned cars, much less cars more generally.
> It's okay to like scooters and bikes, but nothing larger until we get to a bus?
Size has nothing to do with anything (well, it has to do.with efficiency for long-distance transport.)
self driving (they dont even need to be self driving just ride share algorithm powered) shuttles are the future of public transport
The systems could be ride share type shuttles that can pick you up at your location and drop you off at your destination.
You make the request in the app and the shuttle picks people up in a location.
They can even drop you off for a transfer with other modes, other shuttles etc. The shuttle that picks you up doesnt have to be the shuttle that drops you off.
Shuttles like these will greatly reduce traffic and can also handle spikes in demand unlike current mass transit.
Autonomous subways are great for places like Singapore, but there are quite a lot of places where throughput is not an issue and where doubling the mass transit subsidies means hiring a second bus driver so the line can run every two hours instead of every four.
Someone mentioned in HN that a bus every four minutes is a very expensive subway train. I think I agree. I think this is the situation for most New York MTA buses at peak hour. These buses should be in a dedicated track on a subway, not sharing lanes with traffic.
America had more commuter/passenger rail in the 1940/50s than Europe has today. Russia and China are more spread apart and they both have more high speed real than American has.
Every Australian capitol except Darwin/Hobart and ACT have commuter train systems, and they have the same land mass as the US with 1/6 of the people.
I hate hearing people use the argument you used, because it's just straight up wrong. If American built rail, you'd be surprised to see how quickly development would flock to build around it. American spread out like it did because the nation moved to Interstates.
> Russia and China are more spread apart and they both have more high speed real than American has.
How do you explain car craze in china. Its the biggest car market in the world and one thats growing rapidly.
So exactly the opposite of what you are suggesting would happen is happening in china.
China comparisons makes no sense. Car ownership is not accessible to most chinese they have no option but to build a rail system. It is not because ppl prefer travelling in trains(vs cars) its because they have no other option. You have cause and effect reversed. If everyone in china can afford a car like USA they would prefer American interstate system. Who would possibly want to live in cramped quarters built around rail system when you can live in suburban mansion with basement home theatre system.
> Who would possibly want to live in cramped quarters built around rail system when you can live in suburban mansion with basement home theatre system.
Sorry if I sound selfish but it is really scary to think of 1.5 Billion people doing what the 330 million US Americans do.
At some point, I'd argue there is no such thing as sovereignty. I, a neighbor, have a right to not suffocate to death more than you have a right to burn your house to the ground (or I'll take it by force even if I don't have that right).
Is China really more spread apart? It’s a smaller country with a vastly larger population. That population is heavily weighted toward the eastern half of the country, too.
Edit: Wikipedia’s article on China says, “Covering approximately 9,600,000 square kilometers (3,700,000 sq mi), it is the third- or fourth-largest country by total area, depending on the source consulted.“ Now I’m just confused. Area of a country seems like something there would be broad agreement on.
> fully autonomous subways systems like those in Singapore
Nitpick, but in Singapore only three subway lines are fully autonomous (CC, DT, NE). The remaining two are not (EW, NS) because they were built with older technologies.
Most of the money spent on autonomous autos is being spent on research, while most of the money spent on public transportation solves a select few routes.
Solving auto driving essentially solves it everywhere.
Here in Seattle we are spending over 50 billion to put in a couple of light rails, and it won't even be close to being enough public transportation.
The money the government is putting in self driving pales in comparison to what is being put into public transportation.
Even with public transportation, you still need self driving cars. Look at nyc or London, hey both have great subay systems and yet have some of the worst traffic.
Not to mention self driving technologies can help public transportation, freight deliveries, as well as others.
I’d argue that Tokyo’s PT is so good that the last mile is mostly your feet if you’re able bodied. Autonomous cars is mostly something that makes sense in less densely populated places IMO. Which does include most of the US, but frankly I doubt car sharing will make a big impact there soon.
Seattle's ST3 is a massive project and it will solve a lot of problems! The extension goes all the way to the Redmond tech center and the other extension will eventually make it to Everett, meaning there will finally be a train that runs all the time instead of just a few times per day M-F and for select sport events.
There are so many people who cannot afford housing anywhere near Seattle and they are the runs who run your Wal-greens and Starbucks. They have to sit in traffic on I-5 or wait for express buses. This will be a big step up and it's only a start. Seattle needed ST3 a decade ago, and it's going to have a way bigger impact that you think.
> Even if you had a full-autonomous only Interstate where each car could be fully loaded and ride bumper-bumper at over 100kph, it wouldn't even begin reach the carrying capacity of currently existing fully autonomous subways systems like those in Singapore.
This is a poor comparison because cars on highways don't serve at all the same needs as subways in one of the densely populated areas in the world. I'm sure I could construct comparisons where airplanes fail, but a fleet of bicycles succeed(or vice versa), but that doesn't mean that humanity should pick one or the other as the optimal solution.
A full train on the Singapore North East Line can carry 1920 passengers (at 6 passengers/m^2), and at rush hour it runs 1 train per 2 or 3 minutes, so it carries 16 passengers/second.
A fully loaded Toyota Corolla carries five people and is 4.65 meters long, so at 100 km/h each lane of the highway would carry 30 passengers/second. So the highway would carry several times more passengers than the subway.
Even though each train can pack more passengers per square meter than a car, it's not possible to run the trains back-to-back at 100 km/h, because they need to stop to let passengers off and on. At rush hour, the rate of the subway system is basically limited by how quickly people can flow from the platform into the train.
A fully-utilized freeway in the way you describe is one that cars cannot easily get on/off of (where is the space for them?). They will have to wait at the on-ramp to be scheduled into a slot. Because of this, the average throughput of a highway necessarily is less than the max-speed max-packed scenario, since the bottleneck is actually the on/off-ramps and feeder streets.
Of course, but the comment I replied to says that the max-speed max-packed scenario has much less throughput than a subway. That's completely wrong, and it's important to point out that it's wrong in order to see where the actual bottlenecks are (at the on/off ramps and feeder streets).
You're ignoring space between cars, even with full automation you can't have a highway full of cars with no space between them, else there'd be no room for cars to merge on/off the freeway, and any traffic disruption would cause an immediate traffic back-up. Add one car length of distance and that puts the train in the same ballpark as the freeway lane. (I'll ignore the challenge of filling the Corolla with 5 people - I don't want to be the guy stuck in the middle seat)
Though the problem isn't just throughput of the freeway, but also a question of where all of those cars are going to go when they get to the destination -- generally the commute traffic is going the same place at the same time. Self driving cars may not need a parking space, but if they are going to drop off passengers and then drive away to park somewhere outside of the city center, that causes even more city traffic.
because they need to stop to let passengers off and on. At rush hour, the rate of the subway system is basically limited by how quickly people can flow from the platform into the train.
That's why large train stations have multiple tracks.
I have a variation of your claim since I think "true AI" is unfeasible: we will never have fully driverless cars until we build the infrastructure to support them, and in that case, we would have been 100 times better off building trains anyways.
Thanks, I hadn't thought to express it that way. It seems like most of the people optimistic about driverless cars live in the US, and are now trying to deploy them in desert grid-cities like Phoenix. They should try driving almost anywhere else, where the roads have evolved over centuries to support vehicles (cars, but before that carriages) controlled by humans cooperating to get from place to place. Why build robots to operate in that environment, when you can build them their own lane?
He's saying driving is an AI-hard problem. From all the tasks that humans do we don't know which are AI-hard and which can be done by non sapient software, classifying problems is useful.
And I'm saying "AI-hard" isn't a great classification. Is there a known approach to AI that will result in some android who is can learn human languages and human abilities the same way human children do? Is there any reason to think that getting better and better and better at building self-driving cars or computer vision algorithms or any other AI application is going to magically result in some machine gaining "sapience" and somehow having the ability to do unrelated tasks? Is the contention that if someone does develop a self-driving car, that car will also be able to compose an opera or something if we let it? If we do have a self-driving car, should we try and teach it to write poetry or something? Would this car's failure to learn how to compose a sonnet falsify the contention that self-driving cars are "AI-hard"?
To me this all sounds like some type of weird meatbag chauvinism, where it's absolutely impossible to perform any complicated task that a computer hasn't managed to do a good enough job of yet without resorting to building some sort of Commander Data in our image. And, while that may be true for such human abilities as "being able to do a wide variety of genuinely hard things if they are taught, especially at a really young age" or "arguing about abstract concepts we don't fully understand ourselves", driving a car is...well, it's not that hard. Human beings in car cultures learn to do it, but if those same human beings grew up in horse nomad cultures, they would learn to ride horses and shoot arrows from horseback really well, and if they grew up in a different culture they'd get a different skill. Being versatile enough to develop any of those skills given the right childhood experiences seems much harder than any one of those specific skills.
I've got a standing bet with several of my friends that is along those same lines. I haven't gone quite as far as saying it will require true AI, but I did put down money (just dinner, I'm not a big roller :)) saying we're a minimum of 10 years away from being able to buy any kind of level 5 car.
The problems that are reported in the article seems simpler than I imagined. Is google really only gathering data via real world tests? If so then something like NVIDIA simulator will completely crush waymo. https://blogs.nvidia.com/blog/2018/09/12/drive-constellation...
You'd think that simulation is useful for training machine learning algorithms, but in practice, the differences between our best simulations and the real world are such that training on simulations is next to useless.
Google AI blog explains why:
Simulating many years of robotic interaction is quite feasible with modern parallel computing, physics simulation, and rendering technology. Moreover, the resulting data comes with automatically-generated annotations, which is particularly important for tasks where success is hard to infer automatically. The challenge with simulated training is that even the best available simulators do not perfectly capture reality. Models trained purely on synthetic data fail to generalize to the real world, as there is a discrepancy between simulated and real environments, in terms of both visual and physical properties. In fact, the more we increase the fidelity of our simulations, the more effort we have to expend in order to build them, both in terms of implementing complex physical phenomena and in terms of creating the content (e.g., objects, backgrounds) to populate these simulations. This difficulty is compounded by the fact that powerful optimization methods based on deep learning are exceptionally proficient at exploiting simulator flaws: the more powerful the machine learning algorithm, the more likely it is to discover how to "cheat" the simulator to succeed in ways that are infeasible in the real world. The question then becomes: how can a robot utilize simulation to enable it to perform useful tasks in the real world?
The difficulty of transferring simulated experience into the real world is often called the "reality gap." The reality gap is a subtle but important discrepancy between reality and simulation that prevents simulated robotic experience from directly enabling effective real-world performance. Visual perception often constitutes the widest part of the reality gap: while simulated images continue to improve in fidelity, the peculiar and pathological regularities of synthetic pictures, and the wide, unpredictable diversity of real-world images, makes bridging the reality gap particularly difficult when the robot must use vision to perceive the world, as is the case for example in many manipulation tasks.
The simulator isn't intended to fully replace real world training, the reality gap is a well known issue. However it's obvious that with *just real world training that you will never be able to get sufficient data and fail on easily predictable problems as described in the article. A lot of that post is spent on saying how hard it is to build a good simulation, but what is hard for some companies can be much easier for others.
I'm not as pessimistic. I think Google is taking its time since a bad accident would be a PR disaster. Given how big a deal this is, it makes sense for them to take it slowly, and I don't think we can read much into this in regards to the state of their self-driving tech.
I think I agree with this take. Do we really think Waymo cars would be so cautious with merging if a Waymo crash were received with the same level of fanfare and financial repercussions as an ordinary commuter crash? I.e. a $1,000 financial hit and literally nobody cares.
Most of the "mistakes" in the video looked to me more like a hyper-cautious grandmother than an A.I. that fundamentally doesn't know how to drive.
Just an anecdote here, but I live in Phoenix, near their service area, and I see these cars all over the place. I'd say I probably see at least one on each journey I take in my own car.
It seems they have some 'safety level' constant they can control during their training process, so about half a year ago when they were preparing for public release they decided to change safety level constant from 0.99 to 0.999 and something went wrong, like it's completely unable to do some unsafe actions (unprotected left turns, merges, and other tricky situations). And because of that they need additional training with safety drivers for this kind of situations.
so about half a year ago when they were preparing for public release they decided to change safety level constant from 0.99 to 0.999
That's my theory as well. Remember, Uber killed a pedestrian earlier this year. Fortunately that didn't result in self-driving cars being banned or regulated out of existence, but they may not a get a third chance, so it makes sense to be even more cautious after that.
They're just stuck solving the remaining hard cognitive tasks where even humans struggle. They're just not good enough at predicting other cars' trajectory. And nobody really knows at what point they will be.
I'd say we will find the machines to refuse to drive like humans. And we'll have to adapt.
As far as I know, there are no "AI" companies listed on the NASDAQ.
Of course, there are listed companies with significant AI investments. But I'd argue their valuations are driven by their existing/conventional/ad-driven scale, not their AI arms.
Whether or not they're overvalued remains to be seen. The era of cheap money seems (?) to be coming to an end, so it will be an interesting few years.
The poster children of the dotcom boom certainly were - pets.com, webvan, etoys.com (though I can't say with 100% certainty they were all specifically listed on NASDAQ).
It was the IPOs that drove the boom (and subsequent bust). Public investors falling over themselves wanting to get in on the "next big thing". Older generation companies were never going to deliver the kind of quick return multiples that characterized the boom (though as we saw, the enthusiasm certainly jacked up their share prices too, but by and large they managed to ride out the storm).
When i think about the rumours i’ve heard that waymo chief made a 500 millions $ bonus last year... at the time i thought this was because he was the only guy in the world truely capable of transforming the whole industry...
Of course they're not going to unload hundreds of people onto it all at once. Would you?
Of course safety drivers are going to be there for a while. Was the perception really that they would test a bunch and then just pull all of the drivers immediately after? Why would they take such a risk when they can afford to have the drivers monitoring for a while?
Their announcements do seem to be a certain amount of time ahead of reality. They announced the early rider programme having started at least a few weeks/months before the evidence of it started showing up, and it looks the same this time.
My take on that is that it's okay. They're being cautious.
> Was the perception really that they would test a bunch and then just pull all of the drivers immediately after?
If it's after an extended test period, it's not "immediately".
But yes, the perception was that they would test a bunch and eventually pull the drivers out of the typical car. And that they would need to do this before it could be considered a proper launch.
I would think the general climate of fear in the country would make it frightening to launch an ambitious product like this. There's really no hope of a second right now when things go wrong.
Disclosure, I work for Google, but have no connection to Waymo.
I hope the team can keep pushing on; Waymo is one of the most important "moonshots" to the country. It really could save lives. But, one mistake and people won't trust it even though hundreds of automobile accidents happen every day.
I’ll be happy when my TM3 can even recognize stoplights and stop signs on autopilot,
and the adaptive cruise will stop at an intersection and continue on when it turns green. Next they can work on left and right turns.
But in all seriousness, IMO the biggest challenge to AI driving is not object recognition or modeling instantaneous state, but the full hierarchy of “statefullness” which humans keep effortlessly, but which seems to be almost totally missing from the driving AI.
Understanding the current state, the upcoming state, and the expected transitions in between... are crucial to how I am able to turn left with oncoming traffic, smoothly cruise through a 4-way intersection which happens to be on a curve, zipper merge with entering cars coming off an on ramp into my highway lane during rush hour, change lanes to avoid a bad driver, and not slam on the brakes when a stopped car is in the process of turning left up ahead.
When driving on a two lane undivided road, and a left turn lane appears ahead, suddenly an AI finds itself driving in between two lanes and makes a random choice to veer left or right to try to stay in lane. Without the stateful understanding of how and why this new lane came into existence the AI is clueless. The AI can’t solve the problem just by identifying lanes at t=0, or trying to read signs or markings on the road. Us humans know intuitively (or from memory) which lane is for onward travel and which lane is for turning. Any AI driving system built in the next 5-10 years will need an annotated map which tells it which lane to choose.
This all adds up, in my estimation, to a metric fuck-ton if metadata required per mile in order for an AI to operate. It cannot be done “statelessly” (like a human can do it) without an AI at least 5 orders of magnitude “smarter” than what we have today.
It feels like many of these problems could be overcome if every car on the road was also autonomous, and every such car communicated with each other in a network of sources to coordinate traffic, signal important information, etc.
"In another incident, a Waymo car was part of a line of cars approaching an intersection where a car crash was blocking the right lane. The human drivers saw the issue far ahead and began shifting to the left lane. The Waymo car continued straight and only began trying to merge left when it was a few car lengths away from the traffic cones"
So the car actually behaved correctly then? It's called a zipper merge.
205 comments
[ 3.5 ms ] story [ 283 ms ] thread> We now know that Waymo One won't be fully driverless; there will be a driver in the driver's seat.
This reminds me of this famous Radio Yerevan joke:
> Q: Is it true that Ivan Ivanovich Ivanov from Moscow won a car in a lottery?
> A: In principle yes, but:
> it wasn't Ivan Ivanovich Ivanov but Aleksander Aleksandrovich Aleksandrov;
> he is not from Moscow but from Odessa;
> it was not a car but a bicycle;
> he didn't win it, but it was stolen from him.
Because smart people have invested billions of dollars into this technology and because even smarter people work on this technology the general public must pretend that it's all going according to plan, that everything is damn serious and that the emperor still has his clothes on.
"The man who I am about to introduce," he said, "is someone I know you'll enjoy listening to. He is the most gifted businessman in the country. He has made one hundred million dollars in California oil."
The speaker, embarrassed, came to the podium. "Thank you for your kind introduction, Mr. Chairman," he said.
"However, the facts need some clarification. It wasn't oil; it was coal. It wasn't California; it was Pennsylvania. It wasn't one hundred million; it was one hundred thousand. It wasn't me; it was my brother. And he didn't make it; he lost it."
My game leg has been acting up — we’re in for a bitter winter..
Q: Is it true that the great poet Vladimir Mayakovsky committed suicide?
A: Correct, and his last words were “comrades, don’t shoot!”
1. The cost of screwing up is very high.
2. There are important events in the real world environment that are not very recognizable or predictable.
3. Humans accept a level of danger from other human drivers that they absolutely will not accept from robocars.
The challenging cases are probably things like children playing on the side of the road, pedestrians, bicyclists, construction workers, cops standing in the street, people working in manholes, and so on. Relatively small, easily obscured, unpredictable, hard for computers to figure out how they are moving, and very high cost to hit.
Also I think the actual hardware is not good enough yet. LIDAR is great but humans can see really far in really good detail and can work out what is happening even in really bad conditions (rain, looking into the sun, etc).
Non-motorway driving is just too varied. I still think the first real application of autonomous vehicles will be driverless lorries that drive between depots that are connected by motorways. It's way way easier because the environment is so restricted, there's a clear commercial need, and people would be a lot more comfortable about the safety aspect - who hasn't seen a lorry drift into the hard shoulder? Plus lorries are super expensive so it wouldn't increase the cost too much, and they're huge so there is plenty of space for hardware.
If, instead, we have a set of local sensors setup that identified the new "safe" path, an automated car could easily go through this new section. As marked, however, it would be very dangerous to rely on an automated driver.
Speaking as somebody who owns a classic car, drives carefully to baby said classic and often rides shotgun... I gotta say: most drivers are pretty decent. The people I hear complain about how awful everybody else is... they're either old and drive dangerously slow, or they're super entitled and drive aggressively. Since we're on HN, I'm gonna guess you're of the latter camp. Do the world a favor: take Uber and don't write AI.
While it is amusing, doesn't HN encourage people to be civil?
I don't get paid by-the-hour to drive; I get paid to arrive on-time. So commuting in my opinion (like IT in MBAs' opinion) is purely a (time-)cost center. And from that POV the average driver is indeed absolutely terrible. Ever since I started working in the software industry, I don't think I've ever taken the wheel of my car and driven it somewhere, without getting angry at another driver along the way.
Granted, different drivers can be terrible in different ways.
Some of them do stupid mutually-harmful BS like pulling out right front of me with minimal margin and no intent to travel at even half the speed-limit.
Others think that because they have an 8-cylinder engine and a traditional automatic transmission, that they should rule the streets - which they can and do, but that doesn't make me any less angry at them when they e.g. pass me at a 20mph relative-speed on the same side into which I was signalling to change because I was behind a schoolbus.
And no, talking about the avarage human being's ability to operate a specific machine over the course of their life is not painting any individual with a broad brush.
Also, an individual can be a good driver. People in general may not be. That's not a contradiction or an insult. Heck even a "good" driver can drive badly.
Ironically though, Uber is then taking that money and putting it into their own AI so they can get rid of one of their bigger expenses; human drivers.
No, I simply expect others to follow the same laws that I'm bound to and that I follow. The Golden Rule. And a lot of drivers on the road simply do not do that.
> humans are remarkably adaptive and can convey intention through body language -- even through a "puppet body" like a motor vehicle. Humans can, by and large, adapt to "terrible" drivers -- swerving to avoid the aggro asshats
I agree with you here, but this doesn't really help demonstrate that humans are not, on average, poor drivers. That some of us can compensate for the shortcomings of others is great, but the fact that we not only compensate for the shortcomings, but that we're able to suss it out of such fine data — seems only to be evidence that we have a need for it, because collectively our driving is bad. Perhaps you can teach an AI to read the "body language" of a car, but I imagine that such a thing would be remarkably subtle and considerably difficult.
(On the whole, I don't think it really matters what the skill level of humans are. The AI must perform better, or else what's the point? Ideally, we should be able to prove that it's better — so as to sway public opinion, and I suspect that will be the hard part.)
Quite the opposite, on average humans are 99.99999% reliable when driving cars.
Math: https://news.ycombinator.com/item?id=12012166
Computers on the other hand are nowhere near that reliable.
This makes sense, because a lot of times merging into a lane is a game of chicken. You've got to make the drivers in that lane believe that you're going to merge anyway and that they'll need to slow down to avoid hitting you.
A lot of city driving, especially rush hour city driving, is about creating an opening. Not about waiting for one.
But yeah if your turn signal is on and you're not moving into my lane within 1 second, I'm assuming you forgot to turn it off.
Exactly, which is why non-city drivers struggle to drive in big cities. They're used to just waiting for an opening.
I had an accelerated course in aggressive driving over a couple of hours when going through it. In comparison driving in SF or NY is like visiting a spa.
Now, driving in southeast Asia or southern Italy was very, very different from any US city.
I really like this video for describing some of the advantages of "zen driving":
https://www.youtube.com/watch?v=iGFqfTCL2fs
you better slow down and give them an opening or you'll hit them. which you will definitely feel like doing sometimes.
that's one reason i really look forward to the day when cautious, defensive, courteous self-driven cars replace all human drivers here.
Me: it's really sunny today
AI: I love sunny days
Me: me too, maybe that's why I'm so happy today
AI: why do you think you're so happy today?
Asking out of real curiosity, not trying to debate.
What do disabled folks do now in places that don't have such a service?
The only advantage of a self-driving cars is that it might be somewhat cheaper.
Video: https://youtu.be/ArYTxDZzQOM
It sounds like they're taking this seriously.
[1] https://www.youtube.com/watch?v=cdgQpa1pUUE
[2] http://www.visionbeyondsight.org/About-MissionVision.htm
Even if you had a full-autonomous only Interstate where each car could be fully loaded and ride bumper-bumper at over 100kph, it wouldn't even begin reach the carrying capacity of currently existing fully autonomous subways systems like those in Singapore.
https://penguindreams.org/blog/self-driving-cars-will-not-so...
These driverless cars would be great in Europe where they can solve the last leg problems in countries that already have strong public transport infrastructure. In places like America, the existing infrastructure has a long way to go before autonomous cars won't lead to the same gridlock as normal cars .. and an even longer way to go before they're ready for mass consumption.
Citation needed.
> Even if you had a full-autonomous only Interstate where each car could be fully loaded and ride bumper-bumper at over 100kph, it wouldn't even begin reach the carrying capacity of currently existing fully autonomous subways systems like those in Singapore.
Citation needed.
So do roads, slow downs happen when cars get on or off of highways.
Going from a road with a high speed to one with a lower speed, if those roads are at capacity, will always result in a backup. Off ramps serve this purpose of course, they basically convert cars going at 60mph into traffic going 30mph through stop lights or some other sort of flow control, but if the roadway is at capacity (as many major cities are), that off ramp as a flow control mechanism falls apart and cars pile up onto the freeway.
The math is unfortunately independent of self-driving cars. If every car is self driving capacity does increase, cars can self coordinate and some of those stop lights become unnecessary, but if 100 cars want to get onto a road that has room for 20, someone is going to have to wait!
Subways naturally don't have this problem!
Roads also allow for multiple lanes. This allows for somewhat degraded service when something goes wrong; a train stopped on the tracks blocks everything. With multiple lanes, roads can be maintained while still in service. Every subway struggles with maintenance windows.
I'm not saying one is always superior to the other - trains have their niche, but it's a niche that will never be as broad that of autonomous cars.
What I don't understand is why do autonomous self driving cars and subways have to be either or. Why can't we do both? After all, we can't use the subway for ambulances and fire trucks so we still need to build some kind of roads even if we had full confidence in the subways...
Just thinking out loud... Unless we had helicopters for emergencies?
Can we have a town with no paved road? Only sidewalks? That's still paved, right? Can trash collection also take the subway?
Not necessarily.
Rail is cheaper to maintain, and there is a huge opportunity cost of paving over a cities most valuable real estate to make roads.
The math on how large the streets would need to be in downtown Seattle to alleviate traffic end up with a giant % of downtown Seattle being torn down and paved over.
The Urbanist has a good article (https://www.theurbanist.org/2016/10/12/brt-is-not-cheaper-th...) on why light rail is cheaper than a huge freeway.
Rail can also have multiple tracks, it costs more, but amortizes really well.
Not having to intermix pedestrian, bicycle, and cars is another huge win for, well constructed, rail projects. Autonomous cars still have to stop at crosswalks! In urban cores, stop lights seriously cut into average MPH for a trip, and start shifting things into favor of rail.
Moving 100kph continuously bumper to bumper, not having to stop to let passengers on or off, my bet is that the interstate has more throughput than a subway train.
> Please don't post shallow dismissals, especially of other people's work. A good critical comment teaches us something.
Handling 30k passengers per hour is manageable for most rail lines around the world.
Torontos Bloor line does 500,000 daily.
Similarly highways are constrained to their own maximums per lane, and counterintuitively raising the speed limit doesn't help. 1500-1700 vehicles per hour is a fair estimate.
http://buckholztraffic.com/UNF/Data%20Analysis/2%20Lane%20Hi...
Asking for citations might be a shallow dismissal, but stating hypotheses without facts to substantiate them is also shallow.
There are other ways to ask for more information, that open discussion up rather than stiffening into confrontation. Please post in that spirit instead.
So a highway filled with self driving cars zipping along bumper to bumper may not technically be able to carry as much as a fully loaded subway train, but it's so much more useful. Take away the casualties and pollution, and it's as close to perfect as we are likely to get.
Because, to me, it sounds like you just don't want to be uncomfortable, and that's fine, nobody does, but it doesn't make something more useful.
And yes, before you say it, that isn't an everyday occurrence. But jackasses refusing to take their bag off the seat so you can sit down is. As are people who can't be bothered to take a shower, or think that cologne will make up for it. The list goes on. Many people are just unpleasant to be around, and I don't see any problem with that observation.
For me, public transit comfort comes down to a couple things. First, the hard plastic seats make my butt hurt, and standing makes my left foot ache. Sorry, I'm getting old :). Second, the 2x time it takes to get anywhere means I'm on that terrible seat for even longer. And third, because yes, it's absolutely true, I find that perhaps one in ten people is unpleasant enough that I want to be out of visual and/or aromatic range.
Yeah, same for me. And I've heard that the ~80s crop of NYC subway cars were made in Japan. Except there, they have upholstered seats. But that wouldn't work out well in NYC, so they got hard plastic. But on the other hand, the oldest LIRR cars (refurbished for the 1966 World's Fair) had upholstered seats, and they were ~OK.
Similarly, never have I been given religious literature or panhandled. This happens about 90% of the time.
A guy masturbating at the end of a bus.
A stupid kid throwing a can of Vimto to me from the top of the 2story bus.
Car it is for me.
Also people seem to focus exclusively on the negative interactions which is a very human thing to do, but this discounts the positive interactions.
It can be really nice seeing people offer their seats to the elderly, or kids meeting and playing, or the occasional furry companion in a cute bag.
But I enjoy people watching in general. It can be neat to try to imagine a persons world and how it looks.
Except in lucky places with great underground railways you basically have awkward last-mile journeys where you have to take a taxi or hit a car or take a bus and that adds a lot of time to the journey.
It would be great if there were trains everywhere but frankly that's even harder than making a reliable driverless car.
If I have a car, I load everyone and everything up, and we go. We get to our destination and unload, ta-da! Is this not an order of magnitude more useful than taking a train?
Sometimes I think HN is filled with young people with uncomplicated lives :).
No, if these are big issues for you, you likely live in an American low density area where public transportation is scarce. In large metropolitan areas (which is after all, where most of the world already lives, and will live), getting on and off public transport is usually not a problem.
I have relocated to Tokyo for example about a year ago, and for me, a trip to the grocery store is usually done by foot (same is true for other appliances), and multiple forms of public transport are in walking distance.
With all due respect, the number of people who live in cities as dense as Tokyo are the minority, most of us don't.
However, when I think about loading up my young children with all of their stuff, loading me up with everything I need, etc, and then dragging that down the street a quarter mile, then waiting for the bus, boarding it, riding to the transit center, transferring, riding another bus to the light rail station, riding that downtown, and then doing it all again the other direction... well, I'm pretty confident in my original point that the car is a hell of a lot more useful. Sure, you can exist with only the public transport option, but I was not asserting otherwise. But a lot of people on HN seem to think that not only is it possible, but that it is the superior choice.
So, no offense taken, in fact I appreciate you affirming my original comment about so many folks on HN commenting from the viewpoint of someone with a young, uncomplicated life :-).
As it is those with cars push off these negative externalities onto those who choose not to have them, or can't afford them. So your comfort does have a cost, which you don't pay directly.
Also, tour last paragraph just makes you sound pompous and liek you have no intention of having a discussion but instead only wanting to put others with a different point of view down.
I exclusively use public transport, and regularly do both of these without any problems.
I imagine (edit: many) people who use public transportation would agree with you. You now, since they're usually using it out of necessity and not enjoyment.
Right. We get that. It's very obvious that much of the hype around self-driving cars is premised around people who simply reject mass transit outright and therefore believe that somehow, somewhere, they'll get to ride around in a spacious, individual vehicle in the middle of a packed city and so will everyone else and somehow it will all work out because "technology".
Personally I expect self-driving cars to drive a great deal more conservatively than humans as a result, and I think self-driving cars are going to end up demanding a lot more space and buffer on the freeways. And if they don't at first, one day you'll way up and they suddenly all will, because somebody won some lawsuit and a software update was pushed overnight to double the buffers.
I think other drivers will adapt to this and start taking advantage. Think of what happens when you leave a legal following distance in front of you now, in traffic. Now imagine an autonomous car in the same situation. It would probably end up stopped.
And we haven't even considered networking. When the cars are all talking to each and routed together then it's fundamentally a single machine.
The laws and systems will evolve with capabilities rather than getting stuck on supporting the earliest generation of self driving vehicles and norms.
There's still a chance an automated car encounters something that couldn't be spotted / reacted to. For example a deer/kangaroo/whatever jumping right in front of it. If the cars keep their distance, only the first one crashes. If they form a minimum-spacing train, the only message they can send to the following ones is "you're screwed now".
I don't have much experience with deer behaviour, but kangaroos will happily jump right into anything moving. They will even hit the side of your car sometimes (which they should've obviously seen coming)
Subways / trains in Tokyo are faster than driving, same in London.
No stop lights, no intersections, no delays on the road.
Buses aren't faster, but once you go underground, the speed improves dramatically. Even if there is a 2 minute stop once every mile, that is still less time waiting than comparable stop lights in a car.
I think what it comes down to is that if you have a subway system, that will probably beat cars, in the limited area it serves. Once you have to do transfers to buses, it gets much slower. In the places I've visited with subway systems (SF, DC primarily) this is definitely the case.
That's probably US bias, because as far as I can tell by reporting US public transport is generally very bad and far worse in big cities compared to much smaller cities elsewhere.
I look out the window of the train, though, because otherwise I get motion sick. It's lost time to me.
Private point-to-point loses out mostly for long-distance already, short-distance currently either allows last-mile, station overhead, and related concerns to big down shared transportation or suffers from sharing infrastructure optimized for private point-to-point.
Ubiquitous automated (on-demand, not personal cars actually make the station problem easier, since you need less parking at stations, making station footprints smaller, and reducing delay from arrival at station to boarding.
They could also merge with buses: instead of fixed routes and schedules, vehicle sizes and routing are demand driven for efficiency (perhaps allowing you to pay a premium to guarantee a dedicated vehicle for your party.)
Aside from that, I agree with what appears to be your premise -- that we could invent public transport devices that mitigate and perhaps mostly erase the advantage of the car.
But then I'd ask ... why? Why are we supposed to hate cars? It's okay to like scooters and bikes, but nothing larger until we get to a bus? I imagine a future with all sorts of automated cars that never get into an accident and roll silently along emitting not a whiff of noxious fumes, while being big enough to carry just the right amount. If nobody needs to own their own car, then they can call a car that is sized for their exact needs. One passenger or ten. They efficiently get to exactly where they are going in the minimum time possible with minimum effort. What a dream!
That's not quite my premise.
Rather, it is that shared self-driving automobiles (cars through vans through buses) with centralized automated dispatch are the missing element of the mix of public transport devices that, alongside existing options, does that.
> Why are we supposed to hate cars?
Recognizing that public transport has social benefits does not require hating privately owned cars, much less cars more generally.
> It's okay to like scooters and bikes, but nothing larger until we get to a bus?
Size has nothing to do with anything (well, it has to do.with efficiency for long-distance transport.)
The systems could be ride share type shuttles that can pick you up at your location and drop you off at your destination.
You make the request in the app and the shuttle picks people up in a location.
They can even drop you off for a transfer with other modes, other shuttles etc. The shuttle that picks you up doesnt have to be the shuttle that drops you off.
Shuttles like these will greatly reduce traffic and can also handle spikes in demand unlike current mass transit.
Every Australian capitol except Darwin/Hobart and ACT have commuter train systems, and they have the same land mass as the US with 1/6 of the people.
I hate hearing people use the argument you used, because it's just straight up wrong. If American built rail, you'd be surprised to see how quickly development would flock to build around it. American spread out like it did because the nation moved to Interstates.
How do you explain car craze in china. Its the biggest car market in the world and one thats growing rapidly. So exactly the opposite of what you are suggesting would happen is happening in china.
China comparisons makes no sense. Car ownership is not accessible to most chinese they have no option but to build a rail system. It is not because ppl prefer travelling in trains(vs cars) its because they have no other option. You have cause and effect reversed. If everyone in china can afford a car like USA they would prefer American interstate system. Who would possibly want to live in cramped quarters built around rail system when you can live in suburban mansion with basement home theatre system.
Sorry if I sound selfish but it is really scary to think of 1.5 Billion people doing what the 330 million US Americans do.
At some point, I'd argue there is no such thing as sovereignty. I, a neighbor, have a right to not suffocate to death more than you have a right to burn your house to the ground (or I'll take it by force even if I don't have that right).
https://en.m.wikipedia.org/wiki/List_of_countries_and_depend...
Edit: Wikipedia’s article on China says, “Covering approximately 9,600,000 square kilometers (3,700,000 sq mi), it is the third- or fourth-largest country by total area, depending on the source consulted.“ Now I’m just confused. Area of a country seems like something there would be broad agreement on.
Nitpick, but in Singapore only three subway lines are fully autonomous (CC, DT, NE). The remaining two are not (EW, NS) because they were built with older technologies.
There are so many people who cannot afford housing anywhere near Seattle and they are the runs who run your Wal-greens and Starbucks. They have to sit in traffic on I-5 or wait for express buses. This will be a big step up and it's only a start. Seattle needed ST3 a decade ago, and it's going to have a way bigger impact that you think.
This is a poor comparison because cars on highways don't serve at all the same needs as subways in one of the densely populated areas in the world. I'm sure I could construct comparisons where airplanes fail, but a fleet of bicycles succeed(or vice versa), but that doesn't mean that humanity should pick one or the other as the optimal solution.
A fully loaded Toyota Corolla carries five people and is 4.65 meters long, so at 100 km/h each lane of the highway would carry 30 passengers/second. So the highway would carry several times more passengers than the subway.
Even though each train can pack more passengers per square meter than a car, it's not possible to run the trains back-to-back at 100 km/h, because they need to stop to let passengers off and on. At rush hour, the rate of the subway system is basically limited by how quickly people can flow from the platform into the train.
Though the problem isn't just throughput of the freeway, but also a question of where all of those cars are going to go when they get to the destination -- generally the commute traffic is going the same place at the same time. Self driving cars may not need a parking space, but if they are going to drop off passengers and then drive away to park somewhere outside of the city center, that causes even more city traffic.
because they need to stop to let passengers off and on. At rush hour, the rate of the subway system is basically limited by how quickly people can flow from the platform into the train.
That's why large train stations have multiple tracks.
The only thing we might have are dedicated lanes on interstates for driverless cars, like railroad tracks.
In city driving will never happen without AI.
This is a strong claim, but I'm sticking to it.
Able to teach itself anything from unstructured (unprepared) data.
To me this all sounds like some type of weird meatbag chauvinism, where it's absolutely impossible to perform any complicated task that a computer hasn't managed to do a good enough job of yet without resorting to building some sort of Commander Data in our image. And, while that may be true for such human abilities as "being able to do a wide variety of genuinely hard things if they are taught, especially at a really young age" or "arguing about abstract concepts we don't fully understand ourselves", driving a car is...well, it's not that hard. Human beings in car cultures learn to do it, but if those same human beings grew up in horse nomad cultures, they would learn to ride horses and shoot arrows from horseback really well, and if they grew up in a different culture they'd get a different skill. Being versatile enough to develop any of those skills given the right childhood experiences seems much harder than any one of those specific skills.
Google AI blog explains why:
Simulating many years of robotic interaction is quite feasible with modern parallel computing, physics simulation, and rendering technology. Moreover, the resulting data comes with automatically-generated annotations, which is particularly important for tasks where success is hard to infer automatically. The challenge with simulated training is that even the best available simulators do not perfectly capture reality. Models trained purely on synthetic data fail to generalize to the real world, as there is a discrepancy between simulated and real environments, in terms of both visual and physical properties. In fact, the more we increase the fidelity of our simulations, the more effort we have to expend in order to build them, both in terms of implementing complex physical phenomena and in terms of creating the content (e.g., objects, backgrounds) to populate these simulations. This difficulty is compounded by the fact that powerful optimization methods based on deep learning are exceptionally proficient at exploiting simulator flaws: the more powerful the machine learning algorithm, the more likely it is to discover how to "cheat" the simulator to succeed in ways that are infeasible in the real world. The question then becomes: how can a robot utilize simulation to enable it to perform useful tasks in the real world?
The difficulty of transferring simulated experience into the real world is often called the "reality gap." The reality gap is a subtle but important discrepancy between reality and simulation that prevents simulated robotic experience from directly enabling effective real-world performance. Visual perception often constitutes the widest part of the reality gap: while simulated images continue to improve in fidelity, the peculiar and pathological regularities of synthetic pictures, and the wide, unpredictable diversity of real-world images, makes bridging the reality gap particularly difficult when the robot must use vision to perceive the world, as is the case for example in many manipulation tasks.
https://ai.googleblog.com/2017/10/closing-simulation-to-real...
https://blogs.nvidia.com/blog/2018/12/03/physx-high-fidelity...
Most of the "mistakes" in the video looked to me more like a hyper-cautious grandmother than an A.I. that fundamentally doesn't know how to drive.
That's my theory as well. Remember, Uber killed a pedestrian earlier this year. Fortunately that didn't result in self-driving cars being banned or regulated out of existence, but they may not a get a third chance, so it makes sense to be even more cautious after that.
I'd say we will find the machines to refuse to drive like humans. And we'll have to adapt.
https://i.imgur.com/q7whaio.png
Of course, there are listed companies with significant AI investments. But I'd argue their valuations are driven by their existing/conventional/ad-driven scale, not their AI arms.
Whether or not they're overvalued remains to be seen. The era of cheap money seems (?) to be coming to an end, so it will be an interesting few years.
It was the IPOs that drove the boom (and subsequent bust). Public investors falling over themselves wanting to get in on the "next big thing". Older generation companies were never going to deliver the kind of quick return multiples that characterized the boom (though as we saw, the enthusiasm certainly jacked up their share prices too, but by and large they managed to ride out the storm).
Of course safety drivers are going to be there for a while. Was the perception really that they would test a bunch and then just pull all of the drivers immediately after? Why would they take such a risk when they can afford to have the drivers monitoring for a while?
It's almost like the tech press wants Waymo to aggressively pursue a very public failure.
Even one disengagement every 5000 miles without human backup will be perceived as apocalyptic when your fleet drives thousands of trips a day.
Where are those social media posts?
Waymo's PR is getting more and more divorced from reality with every announcement.
My take on that is that it's okay. They're being cautious.
If it's after an extended test period, it's not "immediately".
But yes, the perception was that they would test a bunch and eventually pull the drivers out of the typical car. And that they would need to do this before it could be considered a proper launch.
Disclosure, I work for Google, but have no connection to Waymo.
I hope the team can keep pushing on; Waymo is one of the most important "moonshots" to the country. It really could save lives. But, one mistake and people won't trust it even though hundreds of automobile accidents happen every day.
But in all seriousness, IMO the biggest challenge to AI driving is not object recognition or modeling instantaneous state, but the full hierarchy of “statefullness” which humans keep effortlessly, but which seems to be almost totally missing from the driving AI.
Understanding the current state, the upcoming state, and the expected transitions in between... are crucial to how I am able to turn left with oncoming traffic, smoothly cruise through a 4-way intersection which happens to be on a curve, zipper merge with entering cars coming off an on ramp into my highway lane during rush hour, change lanes to avoid a bad driver, and not slam on the brakes when a stopped car is in the process of turning left up ahead.
When driving on a two lane undivided road, and a left turn lane appears ahead, suddenly an AI finds itself driving in between two lanes and makes a random choice to veer left or right to try to stay in lane. Without the stateful understanding of how and why this new lane came into existence the AI is clueless. The AI can’t solve the problem just by identifying lanes at t=0, or trying to read signs or markings on the road. Us humans know intuitively (or from memory) which lane is for onward travel and which lane is for turning. Any AI driving system built in the next 5-10 years will need an annotated map which tells it which lane to choose.
This all adds up, in my estimation, to a metric fuck-ton if metadata required per mile in order for an AI to operate. It cannot be done “statelessly” (like a human can do it) without an AI at least 5 orders of magnitude “smarter” than what we have today.
So the car actually behaved correctly then? It's called a zipper merge.