How Autonomous Cars' Safety Features Clash With Normal Driving

An anonymous reader writes: Google's autonomous cars have a very good safety record so far — the accidents they've been involved in weren't the software's fault. But that doesn't mean the cars are blending seamlessly into traffic. A NY Times article explains how doing the safest thing sometimes means doing something entirely unexpected to real, human drivers — which itself can lead to dangerous situations. "One Google car, in a test in 2009, couldn't get through a four-way stop because its sensors kept waiting for other (human) drivers to stop completely and let it go. The human drivers kept inching forward, looking for the advantage — paralyzing Google's robot." There are also situations in which the software's behavior may be so incomprehensible to human passengers that they end up turning it off. "In one maneuver, it swerved sharply in a residential neighborhood to avoid a car that was poorly parked, so much so that the Google sensors couldn't tell if it might pull into traffic."

Comments

Best solution:

[Anonymous Coward]

Ban human drivers.

Re:Best solution:

[Anonymous Coward]

Nuke them from orbit, it's the only way.

Re:Best solution:

[Anonymous Coward]

No surprise this is up modded insightful. I'm betting many slashdotters are horrible technologists who assume the world needs to bend to technology. Simply put, that's not the case. If these cars can't handle driving around humans they are not ready for consumption. The fact that they can't properly work with and adapt with humans on the road means that these cars are unsafe. They may be "safe" from the definition of the laws, but they are not safe if they are causing or instigating traffic accidents. It seems it's blind luck that these cars haven't been the clear cut cause of an accident yet.

Re:Best solution:

[operagost]

I too enjoy taking shortcuts instead of fixing things properly.

Re:Best solution:

[tompaulco]

And dogs and cats and raccoons and moose and cows and trees and power poles and snow and rain and potholes and road construction and miscellaneous debris and so on and so on ad infinitum.

Re:Best solution:

[TWX]

More to the point, autonomous cars are currently not the cheaper technology. Any bill that would attempt to force conventional vehicles off of the road would be stillborn, there are far too many automotive enthusiasts that have already made inroads in the other direction (ie, looser emissions testing rules on cars with collectors' insurance) that it literally can not happen. There would also be pushback from those that simply cannot afford new cars and advocacy groups for them; one can buy running cars for less than $1000 on the used market, it will take a decade for there to even be a chance for a used autonomous vehicle to be that cheap, if not even longer.

There have been lots of discussions on attempting to change driver behavior. Those are also nonstarters. People are not going to change how they drive until conditions in the field force them to do so. Hell, we still have idiots driving below the speed limit in the left lane on busy freeways where they're actually posing a safety hazard and where the law actually states that one can be cited for failing to yield and being passed on the right. Most people probably don't even know the rules for what's defined as stopping (ie, remaining still for two seconds where I live) and have no interest in bothering to learn, and the police don't seem inclined to enforce either, so this simply won't change.

The cars are going to have to learn how to adapt to these conditions.

Re:Best solution:

[Anonymous Coward]

I agree. Autonomous cars have the disadvantage of having to follow hard rules, otherwise they run the risk of causing accidents. Ultimately, if all cars follow those hard rules then everyone ends up safer. However, during the period of human/autonomous interaction, the autonomous cars are going to lose since people drive following a much looser rule set.

I had always envisioned the rollout of autonomous cars as something over the course of 20 years or so. During the first 10 years, autonomy would be enabled on the interstates or major roadways, where there is less interaction where a human driver will not follow those hard rules (No stop signs, multiple lanes of traffic all going close to the same speed in the same direction, etc.) Autonomy would be disengaged when exiting these major roadways. These first 10 years would also give time for autonomous vehicle makers to better perfect their rural control, and possibly work towards being able to shrink down the system to be an aftermarket add-on for those legacy cars (And I say this is important because I think you'd also want a system that can be easily upgraded to account for changes in laws or better systems). Additionally, these 10 years are important for autonomous cars to actually start to saturate the market to be available later for those that want to buy a used autonomous car.

After the first 10 years, the autonomous cars are required for interstate travel and allowed in rural/city areas beyond just for testing. By this point, there should be autonomous cars that are available used, and there are still ways to travel outside of the interstates. Then somewhere between 15-20 years, autonomous cars are to be used everywhere.

Re:Best solution:

[grumpyman]

How did we get horses and buggies off the road?

Re:Best solution:

[blackanvil]

We haven't: if you drive around Amish/Mennonite areas, you have to share the road with horse driven buggies and carts on a regular basis. Most big cities also have horse-driven buggies for the tourists and romantics; New York, Philly, Chicago I've seen myself; I'm pretty sure most major cities do nowadays. So we didn't get them off the road, we just relegated them to an ever-decreasing role in transportation.

Re:Best solution:

[Actually, I do RTFA]

Man, where's that antispam idea form letter when you need it.

You're idea for using all autonomou cars will fail because it:

1. Requires the whole world to transition simultaneously.
2. Is cost prohibitive.
3. Ignores legacy systems
4. Requires re-writing human nature

Re:Best solution:

[TWX]

There will be no cheap conversion kits. They would require too many systems to be present in the cars to begin with, would require access to automotive computers that automakers won't provide, and would not be insurable as there would be too much question of liability.

The only car in production today that I could see being converted would be a Tesla, and that's only because they seem to have been pushing in that direction already and have built-in the necessary stuff to make it happen.

Re:Best solution:

[TWX]

And there are still people that love horses, maintain horses, and ride them in public areas as they're fully allowed to do in large portions of the country.

Re:Best solution:

[TWX]

That's what I was thinking too. Way back when I 'dreamed' of the idea I foresaw RFID or other passive-technology built into Botts Dots, or wire embedded in the roadbed that the electronics in the car could find and use to know the road, and then the car's onboard sensors could do obstacle detection, where basically the driver engages the autonomous drive like they currently engage cruise-control, except that they now don't have to steer or brake. That kind of technology would probably work on limited-access freeways where pedestrians and other non-automotive vehicles are prohibited, so long as disabled vehicles have the ability to signal that they're in an error-state so that the autonomous vehicles can give a wide-berth to the shoulder or other obstructed area. The driver still might not be free to sleep or otherwise completely ignore the road, but if the cars could learn their right-of-way from the road and didn't have to worry as much about unpredictable obstructions like humans wandering into the road then such a system should be fairly safe.

Once people got accustomed to that, I could see more inroads to rural highways where there's a greater chance of humans on the roadway or to slower turning vehicles, and from there, tech to work on suburban and later urban navigation. Basically use each density zone as a means to test navigation and avoidance tech before focusing on even denser areas.

Re:Best solution:

[R3d M3rcury]

Then you will also need to ban Cyclists [...]

Not all cyclists. Just people on fixies. Which isn't necessarily a bad thing...

Re:Best solution:

[jafac]

but they are not safe if they are causing or instigating traffic accidents

. . . to include, traffic accidents where no collision occurs. (ie. snarl-ups, traffic misdirection, etc)

Re:Best solution:

[h4ck7h3p14n37]

...the driver engages the autonomous drive like they currently engage cruise-control, except that they now don't have to steer or brake. That kind of technology would probably work on limited-access freeways where pedestrians and other non-automotive vehicles are prohibited,

Mercedes already offers this technology in their cars.

http://techcenter.mercedes-ben...

Vehicles equipped with Distronic Plus and Steering Assist will auto-steer for up to ten seconds with your hands off the wheel. If you're stupid/daring you can even defeat the ten second timeout by taping a soda can to the wheel.

Re:Best solution:

[SvnLyrBrto]

Once the technology is mature, the difference in insurance rates will probably be significant enough to remove most non-automated cars from the road, even if the initial outlay is still higher. Even in the experimental stage, the safety record of the Google cars is nothing short of fantastic. And the vast bulk of accidents (and therefore insurance claims) are the result of human error or carelessness. If you take human error out of the picture, and the accident rate (and therefore the insurance payout rate) goes down significantly, the insurance companies will almost certainly set rates such that there is a strong incentive to upgrade to driverless cars (or a strong penalty in the other direction).

Re:Best solution:

[TWX]

I want the system to simply work, not to have the security defeated by the vehicle operator to kludge it to work.

Re:Best solution:

[TWX]

Prohibitions are generally only on fairly urban limited-access freeways. There are usually no prohibitions on such transportation on rural highways, even the limited-access type. Same rules govern long-distance running on highways, bicycling on highways, and other non-automotive use.

I suspect that the prohibitions in urban areas are only enforceable because there are other avenues, pun intended, of getting around without using the limited-access freeways.

Re:Best solution:

[HornWumpus]

Just wait till Musk decides 'it's time'. Tesla owners are in for one hell of a ride.

Re:Best solution:

[mr_mischief]

You're not really a Houstonian until you've run over a ladder or dodged two.

Re:Best solution:

[TWX]

You mean, when altruistic visionary Elon Musk enables self-drive so that the cars now can chauffeur the passengers around and can then go self-park unassisted to make door-service really slick for the car owner, or when evil overlord Elon Musk activates the feature that makes all Teslas start attacking the population a'la The Racer?

Re:Best solution:

[HornWumpus]

Option 2.

Re:Best solution:

[sjames]

Attrition. As cars got cheaper and horses more expensive and less practical, the transition happened. Other than interstates, horses are still perfectly legal on the roads most everywhere. It even still happens in Pennsylvania.

Re:Best solution:

[RespekMyAthorati]

the safety record of the Google cars is nothing short of fantastic.

Says who? Google? Of course, they are completely unbiased.

Thing is, a universally accepted test of driving competence already exists: it's called getting a driver's license.
As soon as a robocar exists that can pass the standard driving test given in all states and provinces, day or night, and in all weather conditions, then I will believe that its safety record is "nothing short of fantastic".

Re:Best solution:

[dl_sledding]

To supplement your idea (which is logical and a very positive progression for this technology): this technology is in it's infancy. Google talks about the number of miles driven autonomously; that number is infinitesimal compared to the number of miles driven daily by humans. The start of your 20 year period is, IMHO, a minimum of 10 years away. The overcorrection issues discussed are simply a programming glitch; however, refining that behavior (if you can call it that?) will take time to polish and perfect.

Note that none of the autonomy has been at fault in any traffic accidents. That's a great record and a great start. It's also a very limited set of situations, as there aren't many of them out there. When they become more common and are put into less controlled situations, there may be more blame placed on their inability to adjust to circumstances and the strict adherence to rules. Even though, logically, strict adherence will be the legal requirement, but may not be the proper and safest response in reality.

It will be an interesting time.

Re:Best solution:

[toddestan]

The problem is even if production of manually driven cars stopped tomorrow, through normal attrition even in 20 years you'd probably still have 10% of the cars being manually driven, which would be enough that the computer driven cars will encounter them constantly. And you'd probably never truly get rid of them, just as you still have horses and buggies on the road too (which is something else autonomous cars will have to deal with, even if they are somewhat rare to encounter).

Poor example

[fred911]

"One Google car, in a test in 2009,..."

One would think that in 6 years some improvements would have been made. Do we have a more current example?

Re:Poor example

[Mr D from 63]

It would be good to see what progress is being made, but it will be always be a challenge to have these control systems anticipate what human drivers intend to do.

Re:Poor example

[wstrucke]

"One Google car, in a test in 2009,..."

One would think that in 6 years some improvements would have been made. Do we have a more current example?

It mentions further down in the article that that particular example has already been corrected.

... For instance, at four-way stops, the program lets the car inch forward, as the rest of us might, asserting its turn while looking for signs that it is being allowed to go.

Re:Poor example

[Braedley]

As it turns out, we do. A Google Self Driving Car and a cyclist on a fixed gear bike met at a 4-way stop. The cyclist was doing a track stand (staying upright on the peddles, sometimes peddling backwards and forwards a small ammount) instead of balancing on a foot. This caused the Google car to think the cyclist was going to enter the intersection after the car had started moving, causing it to stop and "wait" for the cyclist, which by this point had "stopped", which the car took to mean that he (the cyclist) was waiting for the car to go (which was actually the case), and so the car would start moving again until the cyclist started his next forward motion to balance himself.

Re:Poor example

[PolygamousRanchKid]

it will be always be a challenge to have these control systems anticipate what human drivers intend to do.

This is complicated by the fact that some human drivers do not even know themselves, what they intend to do. So how should a computer control system be able to anticipate what a human driver intends to do, when the human drivers don't even know themselves?

I really don't think it is that many . . . maybe only 1% of all human drivers. However, one clueless driver can confuse and tie up 99 drivers who know where they want to go, and can communicate it to other drivers.

It's like being on a escalator at the airport or train station. Two folks don't know where they are going. So they stop dead in their tracks at the end of the escalator, blocking the path for all the other folks on the escalator. An accordion affect ensues, with all the folks on the escalator getting squished together. The two people doing the blocking, are totally oblivious to this fact. Their field of vision ends at their own noses. They are entirely engulfed in themselves, and can't even conceive that there are other living beings around them.

This is what happens on the road, as well. The driver of the car parked halfway into the street, is just not capable of thinking, that other drivers might be confused by this. Is the car really parked? Or is the driver trying to park? Or maybe trying to drive away . . . ? At any rate, some drivers need to be taught that it is terribly important to anticipate how others might interpret their actions.

Re:Poor example

[Albanach]

So the earlier example with a car doing much the same has been corrected. Now they have data that shows a bike can do something similar at an intersection. I imagine it will be pretty trivial to produce code that lets the car progress through the intersection slowly, while watching the bike to make sure it stays within a box.

One thing's for sure, a car that refuses to go until the cyclist stops moving or takes their turn is hardly creating a dangerous situation.

It does raise another interesting point though. What is it that US road designers have with four way stops? They place them everywhere, while the rest of the world happily gives bigger roads priority and use yields to allow traffic from side roads to merge. If the roads are very similar in traffic volume, use a roundabout.

Re:Poor example

[AmiMoJo]

The NYT hates new car tech, especially EVs and robots. I seems to be in the pocket of some big vested interests (oil presumably, maybe other auto manufacturers who are falling behind). Remember the infamous Tesla Model S review by that Broder guy, where he did everything in his power to make it fail, exceeding the speed limit and slow-cooking himself with the heater etc.

This is just another hit-piece against autonomous cars. It might even be out to trash Tesla again, since they are introducing autopilot.

Re:Poor example

[Anonymous Coward]

That's not an idiot cyclist, that's a talented one. And he's not "going back and forth" but rather "rocking back and forth" slightly to help stay upright. By not putting his foot down, the cyclist can start moving faster than traditional cyclists who have to lift their foot up before they can begin the process of creating forward movement.

Re:Poor example

[Elros]

It does raise another interesting point though. What is it that US road designers have with four way stops? They place them everywhere, while the rest of the world happily gives bigger roads priority and use yields to allow traffic from side roads to merge. If the roads are very similar in traffic volume, use a roundabout.

In the US, a lot of roads in residential areas were built with no signage whatsoever. The rule at an intersection with no signs is equivalent to a 4-way yield. Basically the right-of-way rules of a 4-way stop with out the requirement to stop even when nobody else is around. However, that leads to accident-prone behavior. People tend to just drive through assuming anyone from the other directions is yielding to them.

With out a clear "main" and "side" street, a lot of cities then added 4-way stop signs. They are relatively cheap and solve the problem effectively. Putting in a roundabout uses a lot more space for intersections that rarely have more than 2 cars meet. It also is a lot more costly than just adding 4 signs.

I will grant you that seeing a 4-way stop at an intersection that regularly sees a lot of traffic is a problem. So is a 4-way stop where each direction has two through lanes and a turn lane for each left and right. Many areas of the US are in fact replacing 4-way stops in high-traffic intersections with roundabouts or lights.

Re:Poor example

[TheDarkMaster]

I see this kind of assholes every day on the street. They walk as if they were the only living beings existing on the sidewalk or in the mall, completely ignorant of what happens around them. I wonder what kind of serious defect these people have to act like that.

Re:Poor example

[0123456]

Putting in a roundabout uses a lot more space for intersections that rarely have more than 2 cars meet. It also is a lot more costly than just adding 4 signs.

Uh, you do realize that, in situations like that in Europe, they just paint a circle on the road and put up roundabout signs, right?

Re:Poor example

[Applehu Akbar]

Arizona is replacing its 4-way stops and light-use intersections with "modern roundabouts" that do not use any more land than is already dedicated. THis saves energy and keeps residential neighborhoods quieter.

Re:Poor example

[Ol Olsoc]

Do we have a more current example?

That would kinda mess up the story now wouldn't it?

This is just a techno FUD story for people who can't stand Google or self driving vehicles to point to and yell "See, SEE? I told you these things will never work!"

Theat they have to point to 6 year old data is sort of telling.

Re:Poor example

[jbengt]

It's like being on a escalator at the airport or train station. Two folks don't know where they are going. So they stop dead in their tracks at the end of the escalator, blocking the path for all the other folks on the escalator.

That is a well known problem in architectural design. Give a clear path for people to exit and clear escalators and the like, and place directional signs where people have space to stop to read them. There's an art to finding a way to entice people away from stairs and escalators after they exit them.

Re:Poor example

[Anonymous Coward]

Ever drive behind a Google car? The cars have surely improved, but if you get near a school zone guess what speed the car drives even when school is out of session? Google causes an accordion effect in those zones, which there are plenty of in Mountain View and Palo Alto (where coincidentally you will find lots of self driving cars.)

Re:Poor example

[jbengt]

The roundabout they built to replace a 4-way stop nearby my house may let you slow down to 5 mph rather than stopping and eliminated the odd drunk driver blowing through the stop sign, but it is more difficult to maneuver than stop signs, and has also resulted in more crashes.

Re:Poor example

[TheDarkMaster]

And here comes a "superdriver", AC of course... Boy, keep in mind that the Google cars are EXPERIMENTAL, that is, for very good reason they can not go around flying low as you think is "normal". And to be honest, if you are one of those who can not bear to drive slowly behind another car when it is necessary (lack of patience and lack of respect for others), you should not even be behind the wheel.

Re:Poor example

[tompaulco]

If the roads are very similar in traffic volume, use a roundabout.

I've seen places where they put in roundabouts for existing road intersections and the problem is that the roundabouts, despite taking about 20 times more space than a normal intersection, was still far too small for traffic to go around safely.

Re:Poor example

[McWilde]

Take your idiotic talent to the track. In traffic, you put your foot down at the intersection. This causes you no delay at all, since you can start creating forward movement with the other foot still on the pedal.

Re:Poor example

[Anonymous Coward]

Strange... When I biked on a regular basis, I could start moving *while* lifting the foot I had planted on the pavement while stopped.

You do it by putting your foot down on the side where the pedal is furthest down, or (if you [sub]consciously favor one foot over the other for starts) by spinning the pedals so that the non-resting foot is up. There's no need to create confusing signals for drivers.

Re:Poor example

[tompaulco]

To be fair, if some idiot cyclist was going back on forth at an intersection I would hesitate to drive as well. Of course, as a human I would quickly lose patience and just start driving with the assumption that he would just stop.

Well, I haven't really met up with a cyclist doing a track stand. In 99.9% of cases, the cyclist just blows through the intersection. The other 1 out of 1,000 times, the cyclist will do circles or figure eights.
Either of those cases will likely confuse the software. It certainly confuses regular drivers, and pisses them off.

Re:Poor example

[hawkinspeter]

In the UK, traffic from the right has priority at roundabouts.

Re: Poor example

[TheDarkMaster]

Hummm... The official rules or your "rules of the road"? For your information in my country you must (by law) drive slowly when you are in a school zone, no matter if she is deactivated or not. More than a smart ass here tried to use your argument and as a result they took heavy fines

Re:Poor example

[Boronx]

It's not a traffic circle without a million dollar island complete with shrubbery.

Re:Poor example

[aaarrrgggh]

That is equally a problem with human drivers not understanding that cyclists often cannot see their gestures behind the glare or tint of a windshield/windows.

Re: Poor example

[Coren22]

If there are 1200 people in a room with a single doorway, that is a fire hazard.

Re:Poor example

[aaarrrgggh]

It is an issue with clipless pedals and not just fixed gear bikes. Putting the foot down kills balance and requires substantially more starting torque.

Re: Poor example

[MitchDev]

You live in a shitty country then

Re: Poor example

[MitchDev]

That's when you strong arm the idiot out of the way and walk through

Re: Poor example

[Blymie]

In Quebec, Ontario and BC (places I've lived), school zones have speed restrictions, but only when defined. In Quebec, for example, that's 8am to 5pm, Monday to Friday, September to June. All to 30kmph signs say just this, too.

And here, travelling more than 15kmph under the speed limit is ticketable too.

Re:Poor example

[ArsenneLupin]

perhaps with `reduce' tickets of say $5, that they could pay out of their google wallet account.

Excellent idea! Maybe, as the car doing the ticketing is google, and as the wallet is google as well, here's an excellent opportunity here to streamline the procedure: Google just takes the $5 fine out of the infringers wallet directly, keeps the $4 commission, and forwards the remaining $1 to police.

Re:Poor example

[Calydor]

Not necessarily useless.

In the school zone, let's say 80% of the kids need to cross the road. They then spread out from there, and just the next road over maybe only 20% of them need to cross, if they aren't staying in the school zone to wait for a bus at which point perhaps they won't be crossing the road at all.

The school zone says that a LOT of children are likely to be present, and extra attention and lower speed becomes required.

Re:Poor example

[rgmoore]

it will be always be a challenge to have these control systems anticipate what human drivers intend to do.

It would help if more human drivers used their turn signals regularly. There's a device specifically designed to let other drivers know what you're intending to do, but most drivers refuse to use it.

Re:Poor example

[ArsenneLupin]

This does, however, break down when cars arrive from all directions at the same time, a situation often used by software developers as an example of a deadlock. But in every other situation, it is clearly defined who goes first, with no signs needed.

Actually, 3 cars (on a 4-direction crossroads) is enough to make the system deadlock, as long as the rightmost car wants to turn left (having to wait for the left-most car whose path it would be crossing). Left-most waits for middle, and middle waits for right-most. So, it doesn't take cars from all directions to make the system break down.

But usually (with human drivers) you solve the situation with hand signs. Three google cars meeting at such a crossroads would wait forever.

Re:Poor example

[Jeremi]

I imagine that most people occasionally get into this sort of Mexican standoff with a car, in which neither driver is quite sure whether the other driver is going to go first or is planning to wait and go second.

When two humans are involved, the standoff can be quickly resolved with a hand gesture (not the one-fingered kind; I mean either a "thank you" wave if you want to go first, or a "go ahead" motion if you want them to)

I wonder if it would be worthwhile to put some kind of display on the front of the autonomous car that would allow it to communicate something similar? There is no reason (other than minimizing cost) for a self-driving car to be inscrutable.

Re:Poor example

[ArsenneLupin]

That all depends on what you consider pissing off the drivers behind the Google car to be.

If anything, those drivers would (should) be pissed off at the cyclist, and not at the poor confused google car. Any cyclist figuring out what's going on (and this guy did, as seen in his blog post) would either put his foot down, or else just go ahead.

Re:Poor example

[ArsenneLupin]

And he's not "going back and forth" but rather "rocking back and forth" slightly to help stay upright.

I'm not sure about the US, but here in Europe the rules of the road clearly say that cyclists have to put down their foot on the road at a stop sign, or else it is considered as a rolling stop. So even if the guy is talented enough to stop with his feet in the air, it would still be against the rules.

Re:Poor example

[CauseBy]

"you are one of those who can not bear to drive slowly behind another car when it is necessary"

You should re-read his comment then reply honestly.

His scenario is when you can not bear to drive slowly behind another car when it is unnecessary.

Re: Poor example

[CauseBy]

Wait, are you claiming to live in a place where school speed zones ("she"?) have the ability to be deactivated, and yet you must obey the lower limit even when it is deactivated? Do I have that right?

What, then, does it mean for the zone to be "deactivated" in your country?

Here in the USA, our signs say "25 MPH WHEN CHILDREN PRESENT" or "25 MPH WHEN LIGHT FLASHING". If children aren't present, or the light isn't flashing, then the limit does no apply. People who drive slowly through those zones anyway are considered unencumbered assbags.

Re: Poor example

[CauseBy]

"And here, travelling more than 15kmph under the speed limit is ticketable too."

Yeah, technically that can possibly happen in the United States but it is very rare. And the problem is that speed limits themselves are stupid low, so if the highway says 55 MPH then a person would have to be going 40 on a damn highway where everyone else is going 85 to get a ticket. The result is that we have these few drivers who are creating very unsafe highway conditions but never get ticketed for it.

What I want is simply enforcement of the rule "slower traffic keep right". That's enough for me. No matter what speed you are going -- under at or over the limit, whatever it doesn't matter -- if the lane to your right is empty or moving faster than you then you should get a ticket.

Re: Poor example

[CauseBy]

"Oh, excuse me sir, could we all get past please?"

Re:Poor example

[CauseBy]

I lived in Boston for a couple years. A few weeks before I moved out I was driving on a surface street and the car in front of me signaled his upcoming turn. I swear this is true; in my brain, my inner monologue went like this:

"What the hell is that weird flashing orange light on that car?"
"Is it a hazard light, but the other bulb is burned out?"
"Maybe I should slow down, something could be seriously wrong."
"You idiot, that's a turn signal."
"Oh, gosh, you're right. I haven't seen one of those in a long time."

That only took a second or two to process, but I was genuinely surprised.

Elsewhere in the country where I've lived, however, I find turn signals to be used mostly consistently. Here in Madison drivers are very good about them. I think it's cultural and geographically heterogeneous.

Re:Poor example

[CauseBy]

Not everywhere. In France where I visited the rule is "yield to the right". It's like an implied four-way yield but it forces drivers to slow way down at every single intersection to see around the corner and check for cars to the right. All that instead of just telling two of the ways to yield.

Re:Poor example

[dunkelfalke]

It is still not a big deal. Especially when using clipless pedals it is far safer to clip out and stand on one leg than trying to keep balance, because if anything goes wrong you'll fall on the side and probably won't be able to clip out during the fall.

Re:Poor example

[CauseBy]

Interesting, thank you. Here's more information.

Re:Poor example

[Elros]

Putting in a roundabout uses a lot more space for intersections that rarely have more than 2 cars meet. It also is a lot more costly than just adding 4 signs.

Uh, you do realize that, in situations like that in Europe, they just paint a circle on the road and put up roundabout signs, right?

So, I've never been to Europe, so I wasn't aware of that. In a few intersections near me that might work.

For most in my area though, that isn't a workable solution. The intersections are small enough that some larger cars can't do a 360 turn with out either jumping a curb or backing up and making it at least a 3-point turn. A large truck (think moving truck or school bus) would have no chance. Moving the curbs to expand the intersection requires moving utility lines and in some cases getting dangerously close to existing structures. For low-traffic roads it's just not worth it.

Re: Poor example

[dpidcoe]

I am all for drivers being polite, but they should not confuse other drivers with breaking right-of-way rules. A lot of people seem to not understand 4 way stop rules as it is, without people throwing a wrench I the whole process.

Not only is it inconvenient, but it's also a hazard and a liability risk. Other drivers who aren't necessarily in a position see the exchange of body language and waves yielding right of way when none was present are expecting events to happen in a certain way and get confused if they don't. And if, for example, a pedestrian in a crosswalk waves at you to go when they had right of way but then has second thoughts and jumps out in front of you, guess who's at fault when the police investigate.

Re: Poor example

[Oligonicella]

Pfft! Bullshit. I just walk into them.

Re: Poor example

[TheDarkMaster]

Oh boy, another one with interpretation difficulties... To the point: The school zones here CAN NOT be deactivated, period. The problem is that always appears an idiot saying it can go running in the area because he would only need to drive slowly on the departure time of children, but at least in my country that is not true and for good reason (for one, you will never be sure you will not have children walking on the street at any time). Try to stop at 60km/h to avoid hitting a child and you WILL discover why you should drive slow in this situation.

Re:Poor example

[fisted]

with a hand gesture (not the one-fingered kind [...])

Oh that one works pretty well, too, no matter whether you or them go first.

Re: Poor example

[Oligonicella]

Quoting you both times:

For your information in my country you must (by law) drive slowly when you are in a school zone, no matter if she is deactivated or not.

The school zones here CAN NOT be deactivated, period.

Perhaps now you can understand the poster's "interpretation difficulties" didn't start with him/her.

Re: Poor example

[TheDarkMaster]

Interesting... So I need to be speeding at 85 because "everyone is doing that", when the speed limit of the road is 55? present this argument to the police of my country and they will respond by seizing your driver's license and towing your car, at best.

Re:Poor example

[erapert]

The NYT is a very conservative journal.

Please stop, people in the office are looking at me funny.

Re: Poor example

[TheDarkMaster]

Oh, a "smartass". If you want a version for dummies, it is "you think school areas always have set time to be effective, but here is not that way." I beg a thousand pardons if English is too crap to properly represent a complex sentence thought out and written in Portuguese.

Re: Poor example

[Calydor]

So, if the speed limit is 55 MPH, why is 'everyone' doing 85?

Re:Poor example

[TheDarkMaster]

Necessary or unnecessary, If the driver in front is driving slowly to respect the school zone then how you will drive fast, going over him smartass?

Re:Poor example

[tlhIngan]

As it turns out, we do. A Google Self Driving Car and a cyclist on a fixed gear bike met at a 4-way stop. The cyclist was doing a track stand (staying upright on the peddles, sometimes peddling backwards and forwards a small ammount) instead of balancing on a foot. This caused the Google car to think the cyclist was going to enter the intersection after the car had started moving, causing it to stop and "wait" for the cyclist, which by this point had "stopped", which the car took to mean that he (the cyclist) was waiting for the car to go (which was actually the case), and so the car would start moving again until the cyclist started his next forward motion to balance himself.

Which to be honest, is what should happen.

After all, I come from a place where cyclists routinely blow through stop signs (doesn't matter if it's a 4-way stop, or a regular 2-way stop), red lights are suggestions and basically the rules of the road aren't followed. Any driver who wishes to not get in an accident pretty much anticipates the cyclist to basically play chicken with a car.

Someone doing a track stand would basically halt traffic because all the drivers assume they're going to blow the stop sign, even along a major road.

Re:Poor example

[MMC Monster]

This is complicated by the fact that some human drivers do not even know themselves, what they intend to do. So how should a computer control system be able to anticipate what a human driver intends to do, when the human drivers don't even know themselves?

This is Google we're talking about. If anyone can do it, it's them.

First thought: use facial recognition against their Google Street View database to positively identify the individual. Then pull up all their online activities and generate a psychological profile of how aggressive a driver they are. Then compare that profile to others in the past as well as that individual in the past to predict what they are going to do next.

Re:Poor example

[Calydor]

I was always told that a single flash of the headlights counts as a signal to the car in front of you to go first.

Re:Poor example

[Calydor]

Well, to play devil's advocate, the guy that gets drunk and kills someone with his car is taken off the road for a very long time. The problem is that in the case of self-driving cars, that same guy is driving millions of OTHER cars all the time.

Re: Poor example

[Dragonslicer]

I am all for drivers being polite, but they should not confuse other drivers with breaking right-of-way rules. A lot of people seem to not understand 4 way stop rules as it is, without people throwing a wrench I the whole process.

Sometimes they aren't being as polite as they think. If there aren't any cars behind them, the other driver would have gotten out sooner if they had just kept going. Instead, the other driver has to wait for them to slow down enough to be certain of what they're doing.

Re:Poor example

[kheldan]

Taking away people's choices is never a popular idea which is why there will always be human drivers.

Re:Poor example

[Dragonslicer]

Well, I haven't really met up with a cyclist doing a track stand. In 99.9% of cases, the cyclist just blows through the intersection. The other 1 out of 1,000 times, the cyclist will do circles or figure eights.

Either of those cases will likely confuse the software. It certainly confuses regular drivers, and pisses them off.

You forgot about the pedestrians in the crosswalk that the cyclist came within 6 inches of hitting. It pisses them off, too.

Re: Poor example

[TheDarkMaster]

Hum... You are this dumb? Interesting. Repeat after me: Where is marked as a school zone, drive slow and cautious because children on the road are to be expected. Where is not marked as school zone, drive respecting the local speed limit and rules (or in another words, not a school zone). Is this hard?

Re:Poor example

[morinpatmorin]

I cycle through a residential neighbourhood every day, and I have this problem constantly with _human_ drivers. They simply don't trust a cyclist, who doesn't have at least one foot firmly on the ground, to not pedal out in front of them. I usually have to make a huge "after you" motion with one hand before a car even considers driving through a four way stop in front of me.

Re: Poor example

[Sperbels]

Because the posted speed limit is too slow. Probably to give the local police department a good income from tickets.

Re:Poor example

[Cederic]

It's ok though because we do this innovative thing: We drive over the fucking circle.

Mini-roundabouts: Like roundabouts but you can drive over them.

Re: Poor example

[ColdSam]

Yes, they did start with the reader. A reasonable interpretation of the original statement from this obviously non-native English speaker would have been when "school was not in session." You don't need to automatically leap to the assumption that the poster is an inconsistent idiot. Yet, people do.

Re: Poor example

[ColdSam]

It's well known that in such a room there are indeed a dozen doors, yet people will still all try to go out that one door that is already open.

Re: Poor example

[ColdSam]

But they feel good about themselves and that's the important thing.

Re:Poor example

[ColdSam]

I doubt that's correct. Traffic already IN the roundabout has priority, but when two cars enter a roundabout at the same time neither one has priority because there is no conflict.

Re: Poor example

[Calydor]

Maybe it marks me as a bad driver, but I consider the speed limit to be what people smarter than me in the areas of visibility, road safety and so forth have decided is a safe speed.

Re:Poor example

[hawkinspeter]

In the UK, we drive on the left and go clockwise around roundabouts. If you arrive at a roundabout and there's a car on your right, you're supposed to give way to it unless there's enough room for you to get safely in front of it (without causing them to brake).

Re:Poor example

[The-Ixian]

Yeah, the lack of turn signals always irks me.

When I had a car, I always used turn signals, I would use them even when changing lanes (flip on turn signal to establish intent, check my blind spot, merge to the next lane slowly, flip off my turn signal).

But nobody ever uses them any more. Or if they do, they hit the turn signal as their hand passes by the signal lever as they turn the wheel, which is completely pointless.

Re:Poor example

[Xenna]

Stop pretending that Europe is a country. I never heard of such a ridiculous rule in Holland.

Re:Poor example

[hawkinspeter]

To make it clearer, here's a link to the Highway Code roundabout guidance: http://www.highwaycodeuk.co.uk/news-and-safety-tips/-highway-code-for-roundabouts/

Re:Poor example

[KGIII]

I believe most of those offenses must be personally witnessed by a sworn police officer - one of the reasons that a sworn police officer checks red light cameras. I can not call and tell the cops that you ran a red light. Well, I can, but they can not do anything about it. That is my understanding, at least and only applicable to the US.

Re: Poor example

[Firethorn]

Thus why he said that driving through a deactivated school zone slowly is being an 'assbag'. Because he's not only going slow, he's slowing down everybody behind them.

Look, here in the states residential streets are typically already 25mph - the speed limit for most school zones. Most schools only abut against 1-2 higher speed streets. Get the kids over them safely, the rest of their trip is on 25mph residential streets where kids are to be expected.

We just don't feel the need to keep the speed limit lower when kids are unlikely to be present. Thus the systems of lights. Your system of permanent speeds in this situation is fine, and easier for self-driving cars to handle, but that's not the case in most of the USA and Europe.

Re:Poor example

[KGIII]

Umm... I have tried to avoid commenting on this but it is time. No, just no. A rotary (or a roundabout, if you prefer) takes up approximately the same area as an intersection with turning lanes. They are beautifully efficient and extremely safe until you add uneducated, egotistical, and hurried into the mix. If you drivers were not so stupid we could actually get rid of almost every single reason to stop your vehicle in the road by using a combination of rotaries and yield signs.

Yes, yes, you are most certainly an above average driver and the problem is not with the drivers but with mythical space constraints. (That is a generic you, not you personally.)

No, I did not generally design roads as a career. Worse, my company modeled traffic and told the designers what to do. You guys suck as drivers. And, use Google maps or something. Check the before and after pictures from when an intersection was converted to rotary. It is intuitive to think they take a bunch more space but they really don't. If it's a large group of intersections then you actually end up saving space in some situations. However, it is not conventional space as it is circular.

You move the crosswalks back to the middle of the block (where applicable) and eliminate the turning lane(s) and all that area that they took up is freed up for use. There are some exceptions where a rotary would take up more space and if those areas are poorly designed then there is no easy solution. Using yield signs and educating drivers to treat them like a four way stop should all but eliminate hassles except in the rarest of circumstances.

We don't just make arbitrary choices willy-nilly and put up signage at random. Well, except for Georgia, I think they're drunk. Intersections are replaced with rotaries quite frequently and, inevitably, accidents increase for a short while and then people figure it out and get with the program. Then efficiency increases, accidents decrease, and we can worry about throughput in other areas.

Re: Poor example

[Outta_the_way_peck!]

Because at 86mph they add reckless driving to the list of offenses.

Re:Poor example

[Firethorn]

They simply don't trust a cyclist, who doesn't have at least one foot firmly on the ground, to not pedal out in front of them.

Odd, drivers tend to have that view of me even when I have both feet on the ground straddling my bike, looking at my phone.

I mean, I'm wearing a helmet*, but I'm not in 'racer' gear otherwise.

*Mainly because I've managed to use a bicycle helmet up. I like my skull.

Re: Poor example

[HornWumpus]

In CA you will get a ticket for 'obstructing traffic' if you have more than 4 cars backed up behind you and don't pull over. Doesn't matter if they are speeding or if you are already 5 over. You get the ticket.

That's how it should be everywhere.

Re:Poor example

[HornWumpus]

A-10 cannon under the frame.

Re: Poor example

[HornWumpus]

People who engage in hyperbole are as bad a Nazis.

Re:Poor example

[HornWumpus]

In Boston if you signal it just tells the massholes which way you intend to go, so they can cut you off and flip you the bird.

Re:Poor example

[ColdSam]

I think you're missing the point. Whenever you enter a roadway you ALWAYS yield to traffic already in that roadway. It doesn't matter that it is on the left, the right or wherever, no matter which country. You stop or yield to traffic already there. When you approach a T intersection from the vertical roadway you stop or yield to traffic on the horizontal roadway, even if it is a one way street (left, right? it doesn't matter). Yielding to the right might be a convenient mnemonic device, but I doubt it. It's more confusing then it needs to be.

Re: Poor example

[TheDarkMaster]

"regardless of the speed limit."

You will be jailed here if you think like this...

Re: Poor example

[Kavonte]

> That's how it should be everywhere.

Yes, because the last thing we need is anyone hindering the ability of others to break the law. That needs to be illegal. </sarcasm>

In the more sane parts of the U.S., if you don't like the speed of the person in front of you, you're expected to either pass them or take a different route. Indeed, when I see people following me too closely because they don't like my speed, I help them out by removing my foot from the accelerator until I've slowed sufficiently that they are able to pass me easily. The way I look at it, I can't force people to follow me at a distance that is safe for the speed we are traveling at, but I can force them to follow me at a speed that is safe for the distance they are following me at, and so that's what I do.

Re: Poor example

[guruevi]

Those rules apply in most jurisdictions in the U.S. unless the zone specifically has different signage for it.

I even got a ticket once for going "too fast" (unspecified because 'traffic enforcement officer Donutbelly' didn't have a radar) in a school zone outside active school times.

Re: Poor example

[guruevi]

Speed limits were introduced to curb oil usage in the 70s. They've been a great source of income for jurisdictions since.

Re: Poor example

[guruevi]

Even when not flashing, the school zone speed limits apply as long as there is a sign that indicates a school zone, even in the middle of the night. Unless there is a qualifier that indicates hours, school zones apply at all times in a lot of states.

Re:Poor example

[hawkinspeter]

Sounds like you're back-pedalling there. You stated "I doubt that's correct" and I've provided a link to the Section 185 of the Highway Code that includes "Always give priority to the traffic coming from the right, unless you have been directed otherwise by signs, road markings or traffic lights".

If you think it's just a convenient mnemonic device, maybe you should get the Highway Code re-written to accommodate your views.

Re:Poor example

[tehcyder]

I see this kind of assholes every day on the street. They walk as if they were the only living beings existing on the sidewalk or in the mall, completely ignorant of what happens around them. I wonder what kind of serious defect these people have to act like that.

Being born after 1990.

Re:Poor example

[strikethree]

The whole situation is stupid. You go when it is your turn unless someone is entering the intersection faster than you are. Even then, you should be close to halfway across before you yield to the other person moving out of sequence. Anything else is madness.

If you yield to the slightest of motions, you will never go. If you go when someone else is going more aggressively than you, there will be an accident. The only solution is to assert that it is your turn to go by entering the intersection. If an accident happens despite all of that, then it is clearly the other person's fault and the accident was going to happen anyway.

Re:Poor example

[wstrucke]

With all seriousness, I bet we will see a lot of changes once the number of automated cars reaches a certain threshold. Computer controlled vehicles can do a lot of things safely that we can't.

Re: Poor example

[Stuarticus]

Where you come from kids don't play in the school grounds at funny times of the day?

Re: Poor example

[MitchDev]

Not school playgrounds, no, city parks mostly.

And if the law says the regular speed limit is in effect, then that's what you should be doing, not becoming a traffic hazard by driving much slower than the spped limit

Re: Poor example

[MitchDev]

Most schools don't want kids on their playgrounds out of school hours, legal liability and all

Re: Poor example

[Stuarticus]

After school clubs, sports clubs at the weekends, none of these exist/happen in your limited experience?

Re: Poor example

[Stuarticus]

Most speed limits are intended to be upper limits road and situation allowing, it's actually quite rare to see a minimum speed limit marked. This assumption that you have to do that speed exactly regardless is more likely to make you a hazard.

Re: Poor example

[MitchDev]

They happen, but YOU are the hazard if you aren't obeying traffic laws

Re:Poor example

[ColdSam]

There's no backpedaling, the highway code is written explicitly for morons. I made my point clearly, but perhaps you don't want to acknowledge the inconsistency in your position. Would you say that when entering a T intersection from the bottom that you "yield to traffic on the right if the bar is going left and yield to the left when traffic is going right and yield to both sides when traffic is two-directional?" Of course not. You would just say "yield to traffic already in the roadway." JUST as you would do with a traffic circle. SOME people need a little help with that so you can give them the hint "psst. look right."

Re:Poor example

[hawkinspeter]

Your example misses out the instance when several vehicles arrive at the same time and the roundabout is small enough (e.g. a mini-roundabout - equivalent to a standard T-junction or 4-way junction) that only one car can safely traverse the roundabout at a time.

I just don't understand how you can state that I'm incorrect and then basically repeat the "give way to the right" rule in different language.

Do not bother replying to this as you are just a fool.

Re: Poor example

[Stuarticus]

I think you misunderstand the law or the concept of an upper limit. This may make you a hazard.

Re:Poor example

[Reziac]

They may be fine in Arizona, but they're all kinds of fun to navigate when the road is icy, making every time you turn an invitation to start sliding. Did anyone not notice that roundabouts are a continuous turn? (Which is why I saw lots of accidents at the roundabouts I had to regularly use during an Idaho winter.)

Re:Poor example

[ColdSam]

You're a fool for not attempting to understand the subtle point I was making (and how it relates to writing rules for self-driving cars) and an ass for making this discussion confrontational rather than something that could have been productive.

Of course you yield to the right when entering a roundabout, you just don't need a special rule for it, even if it mentions it in the vehicle code. You also don't need a special rule that says "use your signals in a roundabout" or "watch out for other drivers in a roundabout" which it also says in the vehicle code.

Re: Poor example

[HornWumpus]

If you don't get out of the way, you are the one breaking the more important/serious law.

Re:Poor example

[Ol Olsoc]

It's the same sort of story as when Microsoft commissioned a study to show how Office 365 was less expensive to use than Open Office, using some outfit in Italy that went open source.

http://www.techrepublic.com/ar...

But wait - the Office 365 was compared to 6 year old Open office software, and numbers were fudged a lot, as in while the cost of learning the software was considered the same across the entirety of the study, whereas, as with any software product, it falls over time, from when everyone must learn it, to when only new or inexperienced people do. Nor does it take the costs of changing from OO to Microsoft.

So sorry, 6 years ago, the self driving cars were just starting out, so I'm not going to get all spun up about software issues. After all, that's why they have testing.

But yeah, they had that issue back in 2009. Six years ago. Makes a good talking point for ya.

Here's another one. When dissing solar cells, just the type Charles Fritts invented in 1883 as your metric. It used gold and selenium - and had about 1% efficiency. With your logic, thats what they are all like today...right?

Maybe time moves on, and things improve. If you have to go back six years to find this software problem, it probably doesn't exist any more.

Re: Poor example

[Kavonte]

I don't see the harm in driving down a road at 55 MPH when the speed limit is 55 MPH just because there are four people behind me who want to go 70 MPH.

Indeed, what you describe with having to pull over to allow them to pass would create a situation where those who choose to obey the law on a 55 MPH road are reduced to averaging about 20 MPH, since every time they start moving, four more people with no regard for safety or the law are on their tail again, and so they have to stop yet again to allow them to pass.

Then what happens when you, driving 85 MPH down this road (because that's as fast as you feel you can safely travel) are being followed by a trail of others wanting to drive 115 MPH. So now you have to pull over constantly, reducing your average speed to a mere 40 MPH.

It really makes more sense if the people behind are responsible for resolving the situation on their own by merely passing the slower driver when they have the opportunity. That way everyone can drive at least 55 MPH.

Re: Poor example

[sjames]

There is a school zone near my neighborhood where you cannot see if the lights are flashing or not if you enter it from a side road. You will still be ticketed if the light you had no chance to see was flashing. And yes, sometimes they fail to disable the light for school holidays. In other cases they deliberately leave them enabled to accommodate extracurricular activities. So unless or until the county starts officially publishing when the zones will be active, automated cars and people using that side street are out of luck.

Re:Poor example

[sjames]

Too many drivers here apparently believe yield means ride the bumper of the car ahead so nobody can sneak in. The side streets might not be able to move until the lead car sacrifices itself to a collision or midnight. Meanwhile, the drivers on the side street believe that stopping once when they were 8 cars back was enough so they will tailgate a car that is yielded to so they can also go. For whatever reason, drivers take the stop sign (a little) more seriously so they use those.

You can just imagine the carnage a roundabout might cause given how "well" drivers here handle yielding.

Re:Poor example

[toddestan]

It's common in purely residential neighborhoods around here in the suburbs that normally receive very light traffic (usually because the neighborhoods purposely only connect to one road so there's no through traffic). Typically the rule seems to be that people on the main road have right of way over the people entering from the side roads, most of which are dead ends or finger streets.

Re:Poor example

[toddestan]

Around here, they just need to stop. If he can come to a complete stop without putting his foot down then he's okay. The little dance he did though likely would not count as a full stop.

Re:Poor example

[toddestan]

The problem I have is that many human drivers seem to think of bicyclists as pedestrians, and assume that they pretty much always have right away like pedestrians do. It becomes tricky with the biking trails around here, as bicyclists are supposed to stop and wait at road crossings (the trail has a stop sign and a sign warning that cross traffic does not stop). But about 90% of cars will treat the crossing as a crosswalk and yield to the bikes. Which creates potentially dangerous situations for the 10% of cars that do what is actually the right thing and not stop.

However, I've also found that when I'm on the road, following the rules and acting like a car, people generally treat me like a car. I've had very little trouble at 4 way stops as long as I'm in a lane, stop and wait my turn, and signal if I'm going left or right.

Re: Poor example

[CauseBy]

My guess is that a judge would be sympathetic to the side-street scenario, but the school holiday scenario would depend on the wording of the law.

In other news

[Anonymous Coward]

Millions of people on the road today deserve to have their license taken from them because they can't follow simple rules like signaling, not parking halfway out into the street and leaving enough room to brake in case the car in front of you brakes.

Re:In other news

[Anonymous Coward]

Millions of people on the road today deserve to have their license taken from them because they can't follow simple rules like signaling, not parking halfway out into the street and leaving enough room to brake in case the car in front of you brakes.

Due to this - I am a pedestrian - I'd much rather have self-driving cars.

Re:In other news

[Lumpy]

Or speeding in residential areas. Those people are the scum of the earth.

Re:In other news

[Mr D from 63]

Left lane laggards are even lower.

Re:In other news

[TheDarkMaster]

This. The problem is not who follow the rules, the problem is the idiots who refuse to follow the rules.

Re: In other news

[Anonymous Coward]

I'd think, as a pedestrian, you'd prefer to have self-walking shoes.

Re:In other news

[Anonymous Coward]

Not your job to prevent others from speeding. In fact in most cases the law is written forcing you to yield the left lane to the speeder as they are traveling faster then you.

It is more dangerous for you to go slowly in the left lane and force the speeders to pass you on the right, then it is for you to just get the hell over.

Re:In other news

[internerdj]

While they are indeed a problem, possibly the largest, there are two problems with that being THE problem. 1. Driving safely is often but not always about following the rules. 2. Sometimes the rules are written in a way that happen to counter safe driving to increase the local revenue.

Re: In other news

[Applehu Akbar]

"I'd think, as a pedestrian, you'd prefer to have self-walking shoes."

Ask, and ye shall receive:
http://fortune.com/2015/08/07/...

Re:In other news

[onkelonkel]

We just passed a law basically saying "keep right, except to pass". Judging by comments on news articles about the law, this concept is utterly incomprehensible for a large number of drivers. Any number of folks seemed to think that as long as they were doing the speed limit, this rule didn't apply to them, and that it was ok to stay in the left lane all day.

Re:In other news

[tompaulco]

No, somebody who doesn't enable you to break the law even harder than before isn't lower than somebody who endangers the lives of children. Entitled assholes like you are actually the cause of most problems.

Nice strawman. You assume the person being blocked is breaking the law. That is not necessarily the case. The person being blocked may be traveling at or under the speed limit and legitimately trying to pass traffic going even slower. Meanwhile, the person doing the blocking is absolutely for certain breaking the law because it is illegal to drive in the passing lane unless you are passing somebody. If somebody is able to pass you on the right, then you are not supposed to be in that lane.

Re:In other news

[Derec01]

Not sure where you live, but in my state, the left lane is for passing. If you linger there while not passing or turning, you are technically violating the law. Here's a map: http://jalopnik.com/5501615/le...

Many other iterations of the law specify that you should not block the "normal flow of traffic", specifically distinct from the "speed limit".

Speeding in a residential area can be more dangerous, but you're still often in the wrong if you're doing exactly the speed limit in the left lane.

Re:In other news

[Coren22]

Those people going slow in the left lane are causing just as much of an issue. If someone behind them doesn't realize they are going 20mph below the other traffic, that is a serious accident at highway speeds and will involve many others. On a residential street, yes you are being an asshole, but it is unlikely you will get in a 10 car pileup on a residential street.

Re:In other news

[Coren22]

As a Maryland resident, I am surprised by that map. I have heard of people being pulled over for doing 5 under on the highway in Maryland, and have come across the ones in the left lane pulling off 5 under being passed by everyone. I would expect that it would fall under an impeding the flow type of law.

Re:In other news

[Coren22]

Meanwhile, the person doing the blocking is absolutely for certain breaking the law because it is illegal to drive in the passing lane unless you are passing somebody. If somebody is able to pass you on the right, then you are not supposed to be in that lane.

In 28 states. It is unfortunately not the nationwide law.

Re:In other news

[Coren22]

But typically a lot of speeders are really, really low IQ types. you see them park on someone's rear bumper for miles flashing headlights, as if simply sliding to the right to pass the slowpoke is way too difficult for them..

Because in some states, passing on the right is illegal. I am not able to find a pretty map that shows this one, so you will have to look into if it is legal in your state.

It is however illegal to drive in the left lane (not passing) in 29 states, that is more than half, so it is something people think about.

Re:In other news

[cant_get_a_good_nick]

Oddly enough, me and my wife argue about #1 and #3.

No signaling: if you signal to change lanes, a lot of times that's just a signal to the guy in the lane you're going to... "wait, you're going in front of ME! fuck that" and he hits the gas. I get the "why do you bother, you're just making it harder on yourself to drive.

Safe following distance: my safe following distance tends to be big enough that someone will always jump in front of me.That means i need to now drop back, for another safe following distance. That allows another car to possibly jump in front... Lather, Rinse, Repeat. I get the "do you really want to go slower" from the wife.

So, me trying to drive safely causes me to be much slower because a lot of other jerk drivers. Perverse incentives.

Re:In other news

[Mr D from 63]

We just passed a law basically saying "keep right, except to pass". Judging by comments on news articles about the law, this concept is utterly incomprehensible for a large number of drivers. Any number of folks seemed to think that as long as they were doing the speed limit, this rule didn't apply to them, and that it was ok to stay in the left lane all day.

Glad you did. I think I heard New Jersey was working on a similar law that required you to move out of the way of other cars even if you were doing over the speed limit. So many idiots don't realize how much the screw up traffic and cause clustering of traffic because they just simply don't understand common sense.

Re:In other news

[Mr D from 63]

But typically a lot of speeders are really, really low IQ types. you see them park on someone's rear bumper for miles flashing headlights, as if simply sliding to the right to pass the slowpoke is way too difficult for them..

No, they are trying to reinforce proper driving behavior. The person who is slowing down traffic in the wrong lane needs to pull over, and deserves some negative feedback from other drivers if he/she doesn't. Passing on the right is actually illegal in many places, and is generally less safe.

Re:In other news

[Mr D from 63]

Entitled assholes like you are actually the cause of most problems.

Sounds like you are one of those idiots that does not understand why passing left and keeping right helps everyone. And feels 'self entitled' to slow down others and impede traffic by lagging in the left.

Re:In other news

[Mr D from 63]

Not my job to accommodate them. .

True, your job may be to be a perfect asshole and impede others. But legally, and common sense-wise, and just as common courtesy, get over and drive properly.

Re:In other news

[Mr D from 63]

"I hate tailgaters. The only worse drivers on the road are the ones who drive so slow, you have to ride their rear bumper to get them to speed up." George Carlin

Re:In other news

[Cro Magnon]

Those people going slow in the left lane are causing just as much of an issue. If someone behind them doesn't realize they are going 20mph below the other traffic, that is a serious accident at highway speeds and will involve many others. On a residential street, yes you are being an asshole, but it is unlikely you will get in a 10 car pileup on a residential street.

No, but you could mow down 10 pre-schoolers playing in the street.

Re:In other news

[Cro Magnon]

Using the left hand lane for exits or entrances to interstates is dumb. D. U. M. B.

I agree, but I didn't build the damn things. I just use them.

Re:In other news

[cbhacking]

This is already law in the state I live (Washington), and has been so since before I get my license. It is *almost* never enforced.

With that said, there was a time at a past employer where somebody posted mail to a large social (voluntary inclusion, not-work-related) internal mailing list asking for advice on how to get out of a ticket he'd gotten for holding up traffic in the left lane. The typical response ran something like this:

Well, did you get the cop's name/badge number? Because I'd like to buy him a beer sometime.

Re:In other news

[Coren22]

Do you often let groups of preschoolers play on a thru street? They should be playing in a park.

Also, what kind of oblivious would you have to be to miss 10 preschoolers playing in the street? A single kid running after a ball from behind a car, sure, but a whole group is quite unlikely.

Re:In other news

[Cederic]

Yes. It's because I live in a residential area, where there are children.

Re:In other news

[Cro Magnon]

They're not my kids. For that matter, it's often not on my street (though I have seen kids in the street on my block). Their parents probably let them play in the street because the park is too dangerous. :-P

As for the obliviousness, given the driving I've seen, I wouldn't put it past some drunken idiot to miss, or rather not miss, the kids.

Re:In other news

[chihowa]

If you see this sign and the lane to the right of you is clear enough for people to pass you on the right, then you're not legally in the right by being in the left lane.

Quit bitching and whining about the dangerous behavior of others while you're actively breaking the law and needlessly adding to the hazard of the whole situation. Talk about "zero self awareness of your unaccommodating behavior".

Re:In other news

[Derec01]

If you speed in the left you are also technically violating the law and in many more places than you would be violating it by doing the speed limit in the left lane.

So yes, people going at the speed limit are violating the law in some places (which I never disputed so you can kindly stop pointing it out repeatedly) but they are not particularly dangerous unless mixed with other people violating the law.

When speeding in a residential area the danger created by a single driver. When traveling in the left lane this is rarely the case, ergo you are arguing from personal annoyance than any regards for the danger created by the driver.

I was attempting to be informative; someone stated that it wasn't their job to be "accommodating". Legally, you are absolutely supposed to be accommodating in many states by staying out of the left hand lane when it isn't necessary for passing or turning. Not sure what you mean by "repeatedly", as that was the first time I posted on this issue.

Again, your last point is correct. It's more dangerous to speed in a residential zone. I wasn't taking issue with it.

Very pokey and slow

[Anonymous Coward]

From what I have read from people who have actually interacted with a autonomous car. They are very pokey, slow, and tend to pause trying to figure out what to do.
In fact many times its the required human driver who has to intervene in order to help the car out of a jam. I think the more we try and mix these auto driven vehicles with human one's the more we will experience the growing pains of this technology.

Re:Very pokey and slow

[UncleGizmo]

I see this autonomy being helpful, like an autopilot control. I may be the one in control as I enter the highway, get to speed and merge, but once I'm in a lane, I may switch over to the autonomous system.

Re:Very pokey and slow

[jrumney]

And likely the first iteration of commercially available autonomous driving will be focused on specific easy and low risk tasks like this (you could say that with adaptive cruise control and lane departure warning we are almost there already). Very low speed manoeuvring (autonomous parking) and long distance cruising on dual carriageway roads is the low hanging fruit of autonomous driving. Urban driving is where the hard stuff is, and while Google has made great progress it is still a long way from becoming universal.

culture dependent

[Chrisq]

“They have to learn to be aggressive in the right amount, and the right amount depends on the culture.”

Very true, When holidaying in Texas I quickly found out that stopping for a red light that had just turned would upset drivers behind me. The lights had a much longer amber time, so a whole lot of people who would have had to brake for the lights in the UK would go through

Re:culture dependent

[Pentium100]

Depends on the location. In my country, a yellow light means you can legally drive out of an intersection (for example, if you are turning left), but not into it, unless you need to brake suddenly, in which case you can go.

Now, "brake suddenly" is a subjective thing. If I see the light some distance away and do not need to slam on the brakes to stop, then I stop. It may cause an inattentive driver to hit me from behind (happened recently when I stopped to allow a pedestrian to cross as required by law).

However, it is also the law to leave a safe distance between you and the car in front so you can stop without hitting it if the car in front of you suddenly stops. If you hit another car from behing, you will almost always be found guilty (pretty much the only hope for you is for the other driver to be drunk - drunk drivers are always guilty for an accident even if they did not cause it - this is done to discourage people from driving drunk).

Re:culture dependent

[Malc]

Amber means stop you can do so safely:
https://www.gov.uk/government/...

It does not mean try to squeeze through because you think you have time. I know people who've had tickets for running the amber, despite being across the line before it went red.

Re:culture dependent

[Anonymous Coward]

Turned to what? You aren't supposed to stop the instant the light goes yellow. It's not a stop light. It is just signaling that it is about to turn red, go if braking is not safe.

Well if you're from an area with very short yellow times (*cough*chicago*cough*) then you can easily get into the habit of slowing down for a stop the moment it turns yellow to avoid getting stuck in the middle of a the intersection during the red or get a red light ticket. Then you go to a place with long yellow times and people treat it like a caution light (which is what it is supposed to be) and this causes problems if you suddenly stop instead as if it was already red.

What annoys me more though is the lights that turn red and then after a delay the green right arrow comes on, it makes more sense to have the green arrow come on at the same time so that the right lane doesn't waste time and energy slowing to a stop only to be told to get going again seconds later.

Re:culture dependent

[jabuzz]

The other get out for hitting a driver in front is if said driver pulled into your safe braking distance.

So if some idiot pulls out of a junction without looking and you go into the back of them as a result it is not your fault. Another one would be someone overtaking pulling in and then slamming the brakes on (the last one is often done as part of an insurance fraud).

There are a whole bunch of others as well, though they can be hard to prove if you don't have a dashboard camera.

Re:culture dependent

[Pentium100]

Yes, but because they are very difficult to prove without a dashcam, if you do not have a dashcam, you'd better hope the other guy is drunk or is honest and accepts the responsibility.

Re:culture dependent

[Anonymous Coward]

HA! In New York City, amber means ACCELERATE!!

Re:culture dependent

[thsths]

Yes, you are supposed to stop safely when the light goes amber, unless it is unsafe or impossible to do so before the stop line.

And this is one of the areas where autonomous vehicles will struggle, just as the speed limit. Because it is very hard to justify that autonomous vehicle should speed or run red lights, like most people do. It is the difference between a decision on polity and a decision on an individual case.

I for one am curious how this is going to be solved. I think it will also reveal that following the rules is not always (or maybe not usually) the safe thing to do.

Re:culture dependent

[thsths]

PS: Unsafe is usually interpreted to mean using an acceleration that may not be available to all vehicles. Obviously a sports car on special tyres can stop a lot faster than an old banger or a truck, and and therefore you should never use maximum deceleration (emergency stop).

Whether the driver behind you is paying attention is not your problem. Anybody not paying attention at a traffic light should not be on the road.

Re: culture dependent

[fisted]

I'm color blind, you insensitive clod. It's called fucking the middle grey.

Re:culture dependent

[beelsebob]

Depends entirely on the location. In many (the UK, and Texas for example, which seem relevant to the example), the rule is that you must stop if it's safe to do so. If it's not safe, you may cautiously proceed through the junction.

So in this case, it seems like it was completely safe to stop, and therefore, yes, it effectively was a stop light.

Re:culture dependent

[Maxo-Texas]

I've also seen at least a half dozen horrific accidents from that behavior including running a full red 2 seconds after it changed.

People should slow and prepare to stop at amber lights. It is actively dangerous on feeder roads because the combined blind spot with the other road coming under the freeway. A few years ago, I saw one truck wipe out three other cars including an SUV and break off the traffic light by running a red.

Re:culture dependent

[Anonymous Coward]

Probably depends state to state, also what people teach/do in the real world vs what is legal. In my state I believe yellow means stop if possible to do so safely, but I was taught in driving school that as long as your front tires are in the intersection while the light is yellow you're good. My driving instructor also taught us that if you're turning left and traffic is heavy use the center turn lane as a kind of merge lane, that little maneuver caused me to fail my first driving test even though it is a very common technique.

Re:culture dependent

[Ichijo]

Amber means stop [if] you can do so safely

In the USA, it means nothing more than the light is about to change to red (CVC 21452).

Re:culture dependent

[tompaulco]

I really need to get a dashcam and start posting idiots on the web. I would say that I see one or more idiots enter an intersection on a very stale yellow and leave the intersection on a red at about 50% of intersections. About 1 out of every 5 intersections, I see an idiot enter the intersection on red and continue through on red. About 1 in 20 I see an idiot enter an intersection after it has been read a good 2 seconds.
People turning left seem to really enjoy running red lights. I have had my light be green four 4 or 5 seconds and there are still left turners meandering through the intersection on their red light. Occasionally I miss my cycle completely due to people running red lights.

Re:culture dependent

[Triklyn]

really depends on how fast you're going.

my part of texas, we're looking at a yellow that's a good second or two. and traffic can be anywhere from 30-55mph. i usually just give myself a "if it's not yellow by this point, i'm going through this fucker" line in my mind.

Re:culture dependent

[phantomfive]

"Yellow" means different things depending on the jurisdiction. In California, as long as you are across the line before it turns red, you are good.

Re:culture dependent

[T.E.D.]

Here in Oklahoma it just means the light is about to turn red. If its red, and you are in the intersection for some reason, that's when you can be cited.

Yeah, that rule of thumb about not braking "suddenly" because the light changed is still a good rule of thumb, but legally it has no weight. If I get ticketed, "it was yellow when I entered the intersection", or even "I would have had to slam on the brakes" is no excuse.

Found that one out the hard way.

Re:culture dependent

[T.E.D.]

...and yet another one is if it was dark or within a few minutes of dusk/dawn and you didn't have headlights on. Found that one out the hard way too. My son got hit by another teenaged driver on the way to school, made the mistake of answering "I'm not sure" to questions about whether he had his lights on and what the exact time in the morning was, and ended up at fault. :-(

Re:culture dependent

[T.E.D.]

That's the UK. In Oklahoma (each US state has its own laws), if the light is red and you haven't made it through the intersection yet, that's a violation. Yellow is just a nice warning that you'd better get yourself out of the intersection before red happens.

On the plus side, that's a lot more objective. However, it has the unfortunate side-effect of inducing people to step on the gas in intersections.

Re: culture dependent

[R3d M3rcury]

It's not as common as red-green, but blue-yellow colorblindness exists. Yellow will appear as either a violet or light grey.

Re:culture dependent

[Drethon]

This is why I hate traffic lights on 55MPH roads. If the yellow catches me at the right point I have to stop a lot faster than I like as otherwise the light would be red before I reach the intersection.

Re:culture dependent

[Talderas]

Then there's the people that enter the intersection when it's green but they cannot clear the intersection and consequently end up blocking traffic once it has turned red.

Re:culture dependent

[Pentium100]

Here you probably won't get a ticket for being in an intersection when the light turns red if other cars prevented you from leaving the intersection.

If you want to turn left, you have to drive into the intersection, begin to turn, then stop and wait until everyone who is going straight in the opposite direction has passed. If some of them go into the intersection when the light is yellow or red (breaking the law) you have no other option but to stay in the intersection and wait until the road is clear for you.

Re:culture dependent

[tompaulco]

Then there's the people that enter the intersection when it's green but they cannot clear the intersection and consequently end up blocking traffic once it has turned red.

Yes. If you can't clear the intersection, it is illegal to enter it. I will literally sit at a green light and ignore the people behind me honking, because there is no way for me to proceed through the intersection. Of course, then what happens is the a**holes in the lanes next to me change lanes in the intersection (also illegal) to get into my lane and then proceed to block traffic and not be able to exit the intersection.

Re:culture dependent

[T.E.D.]

Well, I don't know about where you live, but my understanding here is that you technically aren't supposed to go into the intersection at all unless you have the right of way, or there's nobody coming. Yes, most everyone if they are first in line for a left does just what you describe (including quite often me). However, you aren't supposed to.

One way this can hose you is that a lot of intersections have a weight on the left turn lane to detect when people need to turn and give them that protected left at the end of the green cycle. If you're waiting in the middle of the intersection and there's nobody behind you on that weight, you won't get your protected left.

Re:culture dependent

[Pentium100]

In Lithuania you are supposed to go into the middle of the intersection (I guess the reason is that then you need a smaller interval to be able to go), but not if there is already a car there (this part gets ignored by everyone).

And there are no sensors on the ground.

Re:culture dependent

[martinQblank]

"When holidaying in Texas..."

Umm... As an American can I just say that your travel agent sucks?

Re:culture dependent

[toddestan]

Around here, you're not supposed to enter an intersection unless you can also clear it. That means that if traffic is stopped on the other side of the intersection, you stop and wait on your side until there's enough room for you to clear it. That also means for the left turn yield on green situation, you wait until it's clear before you make your turn, though myself and almost everyone will creep into the intersection and wait if it looks like there's going to be a break in the oncoming traffic. Though the people who enter the intersection, then make their turn when the oncoming traffic stops for the red light can be ticketed. As long as you follow the first rule in the sense that you only enter the intersection if you can clear it, you're okay if the light turns red while you're in the intersections so long as it wasn't red when you entered it.

Don't call it normal driving

[Anonymous Coward]

It's not normal to disobey simple rules or park like a retard.

Not normal driving.

[Lumpy]

They are confused by BAD driving. People in general really really suck at driving and a computer will have problems with that.

Re:Not normal driving.

[gstoddart]

Which is always going to be the problem ... because as long as there are human drivers on the road, there will always be cases in which the computer utterly fails.

And any technology future which is predicated on suddenly replacing all drivers with autonomous cars is complete crap and will never actually happen. Because nobody is going to pay for it.

It's the corner cases which will always cause these things to go wrong. And, I'm sorry, but the driver with his right turn signal on who swoops across two lanes and turns left ... or the ones who think they can use the oncoming lane because there's something in their lane ... or who randomly brake because they can see a cat a half mile away ... or cyclists who do crazy and random shit ... or any number of crazy things you can see on a daily basis ... all of these things will create situations in which the autonomous car utterly fails to do the right thing.

As much as people think it will mostly work most of the time, if these things require the driver to constantly monitor it or have to swoop in when the system decides it doesn't know what to do, then the utility of the autonomous car pretty much vanishes.

I just don't see this technology ever becoming widespread or used in the real world, other than by companies trying to prove how awesome it is. Because it's just going to have too many cases which simply don't work, and the occupants will have to be ready to take the controls.

In which case you might as we be driving and actively engaged in the process instead of zoned out and not paying attention. Because the human reaction time is greatly diminished when you're reading the newspaper and suddenly have to take evasive reaction.

Re:Not normal driving.

[DrXym]

It's not just drivers. Sharing the road requires cooperation with drivers, cyclist, pedestrians, traffic cops, workmen and anybody else you may have cause to interact with during a drive. How does a self drive car interact with any of them?

A human can wave the pedestrian across the road at a junction - the self drive car will either sit there like an ass or plough through even if it's obvious to a human drive to yield.

If humans encounter a set of crossroad lights that are out they'll tend to order and arrange passage across the lights in an orderly fashion based on who arrive first and so on. If a cop is there, they'll obey the cop's signals. The autonomous car would just sit their like an ass or nudge out into the junction regardless of the consequences.

At the very least autonomous cars need an unimpaired human to extricate them from situations like this. This puts paid to some of the nonsense that people associate with self drive vehicles - no they're not going to drive somebody home who is asleep / drunk, no they're not going to park themselves down the road and come back when called. It's too easy for them to be confused and too easy for them to be griefed.

Re:Not normal driving.

[Lumpy]

It can be fixed. MAke is significantly more difficult to get and keep a drivers license. In the USA that means moving from "are you breathing" to "are you competent"

Re:Not normal driving.

[dcollins]

Bingo, good post. It's the new flying cars.

Re:Not normal driving.

[dave420]

You just made a bunch of assumptions based on a car attempting to cross an intersection 6 years ago. Just because you aren't clever enough to figure out a set of rules which can solve the problems you mentioned doesn't mean they're impossible. Who cares if you don't see this technology becoming widespread or used in the real world - you clearly can't see past your own hubris.

Re:Not normal driving.

[dave420]

You are making so many assumptions it's not even funny. Just because you can't figure out how to solve these problems doesn't make them unsolvable. At the very least the autonomous cars need good coding to handle these situations, and they seem to be getting just that.

Re:Not normal driving.

[Quirkz]

It's the corner cases which will always cause these things to go wrong. And, I'm sorry, but the driver with his right turn signal on who swoops across two lanes and turns left ... or the ones who think they can use the oncoming lane because there's something in their lane ... or who randomly brake because they can see a cat a half mile away ... or cyclists who do crazy and random shit ... or any number of crazy things you can see on a daily basis ... all of these things will create situations in which the autonomous car utterly fails to do the right thing.

You say with authority that the software fails to do the right thing, but I'm not convinced that's a demonstrated fact. The software has just as much opportunity as a human does to react to these things poorly or sensibly. Many of these problems are probably best addressed by slowing down, edging over, or avoiding the object that shouldn't be where it is or is moving in unanticipated ways. People and software can both do that, or can both fail.

The software may still have an advantage: it's going to see the object immediately, while a person might be startled and panic or might take a while to realize what's going on; the software *knows* where the empty spaces are to maneuver into where a human might or might not be completely aware of all the options; and the software, if not now then probably soon, will have a better reaction time.

Re:Not normal driving.

[Jeremi]

And, I'm sorry, but the driver with his right turn signal on who swoops across two lanes and turns left ... or the ones who think they can use the oncoming lane because there's something in their lane ... or who randomly brake because they can see a cat a half mile away ... or cyclists who do crazy and random shit ... or any number of crazy things you can see on a daily basis ... all of these things will create situations in which the autonomous car utterly fails to do the right thing.

Your pessimistic prediction is based on a misunderstanding of how autonomous cars are programmed.

You're imagining that the autonomous car's programming is based on a legal model -- i.e. that it is assuming that other cars will always follow the law and do the right thing according to the DMV handbook. And you (correctly) infer that a program based on that assumption would often fail in the real world, since other drivers sometimes do crazy or illegal things.

Fortunately, Google's engineers aren't stupid, and they understand that problem also. Which is why they don't program their cars to rely on that assumption. Rather, they program their cars to assume only that other cars will follow only the laws of physics -- for example, they can safely assume that a car will not accelerate from 0mph to 150mph in one second, but they cannot safely assume that a car will stop at a red light.

Approaching the problem at that level is not only more reliable (since unlike the driving code, it's impossible to break the laws of physics) but also easier (since the laws of physics are simpler and easier to quantify than the US driving code, and they are MUCH easier than trying to predict human drivers' psychology). An autonomous car avoiding other cars on a real street isn't much different than a bot avoiding enemy missiles in a video game -- in both cases, the program knows the missiles' current positions, their velocities, and the possible ways in which they could accelerate/decelerate/change-direction if they chose to do so -- and it plans and executes its own actions accordingly. It's not trivial, and it's not guaranteed to avoid every possible accident -- but it's not rocket science, either. Remember that to succeed, the car doesn't have to be God-like in its abilities, only an order of magnitude or so better than the average human driver.

None of this is to say that the autonomous car doesn't need to understand the rules of the road -- but its understanding of the law is used to guide its own actions, not to make assumptions about the actions of the other cars.

Re:Not normal driving.

[DrXym]

Look up what intractable means. I could program a perfect cross-the-broken-lights algorithm that would utterly fail if one of the other vehicles was not autonomous, or was a bus, or if there was a lane out, or if there was a cop in the road directing traffic, or if there is a crazy person directing traffic, or if there is an emergency vehicle coming, or if one direction has natural right of way, or if the conditions are icy, or if the cross roads has some peculiar filter lanes, or if there were people waiting to cross, or, or, or. And that even assumes the car has a way of figuring the lights are out in the first place, as opposed to only being turned on during certain hours of the day, or obscured by snow or whatever.

That's the problem and it's not just a case of "if you can't figure out these problems". The problems are intractable.

Re:Not normal driving.

[phantomfive]

Yeah. Autonomous driving is hard. The special cases are so numerous, even getting 25% of them correct would be an incredible achievement.

Re:Not normal driving.

[T.E.D.]

Which is always going to be the problem ... because as long as there are human drivers on the road, there will always be cases in which the computer utterly fails

More to the point, some folks are liable to learn how to abuse the algorithms. For example, depending on how its written, it might be quite possible to "herd" the google car somewhere it wouldn't otherwise go (eg: into a strip mall, onto the railroad tracks with an oncoming train, etc).

Re:Not normal driving.

[kackle]

It's "corner cases" all the way down... I agree with your post, but you stopped short (I couldn't resist).

It's not just unexpected human behavior on the road, but also unexpected, well, everything. I would say that I see something unusual that significantly affects driving just about every other month. Just off the top of my head I have had to deal with the following in the last few months alone:

-- There is local road construction that closed a two-way road, turning it into one lane/direction at a time, controlled by a human construction flagger. How could a GPS-driven vehicle possibly know what to do there, intentionally driving in the wrong lane for a half mile? What would it think the speed limit is if the signs were obscured? And the human with the "Slow" sign was standing well into our lane, causing us to drive around him. Would the (autonomous vehicle) AV just stop? Further into that single lane there were construction cones cutting our lane down to three quarters of a lane, effectively forcing us to drive along a rough, but paved, shoulder. Would an AV have any clue what to do or would it accidentally ditch itself, utterly blocking traffic until a tow truck came?

-- A highway accident occurred in front of the car in front of me, where the car smashed into the wall head on, perpendicular to traffic flow. The first clue I got that something was amiss was its screeching tires followed by the sideways car entering my peripheral vision. What would an AV do there?

-- More local road construction caused a newly formed temporary dip in the road. The construction people put a sign near it alerting drivers to that fact. I assume the AV would drive over it at the speed limit sense it can't read English.

-- There is a "no turn on red" sign near my house which disallows such turns at certain hours on certain days. Again, can the AV decipher English?

I have been driving for thirty years. And I was learning about the world fifteen years before that. An AV can never be that smart about everything. I am an embedded software engineer, and it always tickles me how many engineers believe they can think of everything in advance. They rarely get called on when they can't. How does the song go, "They let the monkey go and blame the monkey wrench." Here are other scenarios I have encountered; have the AV designers thought of these; will you bet your life on it?

-- A live power wire on the street after a storm.
-- A small, long tree branch, which looks like a live power wire, also lying across the street.
-- Is it true that GPS navigation becomes flaky around tall buildings?
-- I was once caught in a torrential downpour where the road was suddenly and completely invisible. I could only navigate because I knew the roadside landmark's well.
-- I have been run completely off the road three times (usually by cell phone drivers). I saved myself and my car each time by swerving onto the shoulder.
-- There was a raging car fire that I drove past. That he was so intense that it instantly warmed my face as I drove past at 40 miles an hour, even though the windows were rolled up.
-- I swerved violently on the highway to avoid being hit by a drunk from behind who was driving way over the speed limit.
-- I have seen roads completely washed out by rain to where it would be dangerous to attempt travel.
-- I have seen red fireworks shot off by amateurs in their driveways.
-- I have seen tumbleweeds cross the road in front of me.
-- I have seen a half dozen cars in ditches cluing me into the fact that black ice was ahead.
-- I have seen emergency vehicles racing to a scene, some with red lights that were visible over a mile away, and some where their lights were not visible until they were near (sirens from an unmarked car).
-- I drove on a local road that was littered with potholes. After months, I finally hit one so deep, I cracked a rim.
-- An open manhole (someone stole the cover for scrap).

Re:Not normal driving.

[perryizgr8]

I think normal driving IS bad driving. Only very few people drive good.

Re:Not normal driving.

[Lumpy]

USA, unless you are completely inept you get one. Hell my daughter IS completely inept at driving and still got one.

I think the only way to not pass the drivers training class is to bring a gun and shoot at other drivers. Except Detroit, it's part of the required driving test to shoot at targets whine driving, but only holding the pistol sideways.

Programmed behaviour is programmed behaviour.

[ledow]

If the programming makes it jerk the steering away from a stationary hazard rather than, say, detect it earlier and slow down as it approaches, then it's not suitably programmed for coexistence with unexpected stationary hazards (Not even anything to do with human presence! What if that was a cardboard box and it swerved heavily in case that box "pulled out"?).

If it can't make it's way through a junction where the drivers are following the rules, that's bad programming. If it can't make it's way through a junction where other drivers don't come to a complete halt for it, it's not fit to be on the road with other drivers.

If you want a car to co-exist on the road, it has to be treated as a learner driver. If a learner driver swerved at a non-hazard, they would fail. If a learner driver refused to make progress at a junction because the masses didn't open up before it, they would fail. So should an automated car.

Unless - and this is important - you are saying that automated cars should only operate on automated roads where such hazards should never be possible and they are deliberately NOT programmed to take account of such things. Which, in itself, is expensive (separate roads with separate rules with no human drivers), stupid (that's otherwise known as a "train line", and because they can't do anything about it it will hurt more when it does happen), and dangerous (because what happens if a cardboard box blows over the automated road? etc.).

Program to take account of these things, or don't plan on driving on the road. The safety record is exemplary but equally there are only a handful of them and the eyes of the world are on them, and there are still humans behind the wheel, and even by miles travelled each one is probably dwarved by a single long-distance driver over the course of a year - and it's not hard to find a long-distance driver who's not had an accident for years.

If you're going to be on the roads, then you need to be able to take account of all these things, the same as any learner driver. Sure, you didn't hurt anyone by swerving or not pulling out, but equally - in the wording of my first driving test failure - you have "failed to make adequate progress" while driving.

A car sitting on a driveway would have an even better safety record but, in real life, it's still bog-useless compared to a human. Similarly for any automated vehicle that just stops at a junction because it can't pull out, or swerves out of the way of a non-hazard (and potentially weighs up collision with non-hazard vs collision with small child and gets it wrong).

Re:Programmed behaviour is programmed behaviour.

[Cassini2]

If it can't make it's way through a junction where the drivers are following the rules, that's bad programming. If it can't make it's way through a junction where other drivers don't come to a complete halt for it, it's not fit to be on the road with other drivers.

The problem is that people don't follow rules. We follow approximations of the rules. For instance, my driver's handbook described the correct way to deal with yielding at a four-way stop as "yield to the person on the right." For a computer, that's an obvious deadlock situation, or worse - an obvious mistake. If four cars are parked at a four way stop, and each car yields to the car on the right, then (a) a situation could occur where no one goes anywhere, and (b) if the individual cars only pay attention to the person on the right, then they could hit an on-coming car turning left, or the car on the left turning left. People process the "yield to the person on the right" rule into something much more complex.

People use a number of complex behaviours at four-way stops. Firstly, the wave of the hand, or the nod of the head to indicate that you yield to the other driver is an important signal. Secondly, in my jurisdiction, 90% of the four way stops are done on a first-come first-served basis. Lastly, and this is the bit I don't understand, often people yield to the person on the left. The actual system of navigating a four-way stop is much more complex than what an initial computer implementation might be.

Re:Programmed behaviour is programmed behaviour.

[Xyrus]

Computers follow rules. Humans (a.k.a every other asshole on the road) do not.

This is a no win situation. If you program a car to drive safely and follow rules, then it won't be safe on roads because of all the assholes who don't. If you program the car to behave more like an asshole ( a human driver), then it won't be safe since there's a good chance it will make the wrong call. If you program the car to just account for assholes but still drive safely, then it will basically choke in situations like a four way stop in southern California where every other asshole will just muscle or roll their way through the stop.

The long pole in the tent isn't developing an AI capable of driving. It's developing an AI that can deal with assholes.

Re:Programmed behaviour is programmed behaviour.

[swillden]

Program to take account of these things, or don't plan on driving on the road.

Duh.

Technology in development is imperfect. Big surprise. These issues are why Google hasn't yet started selling them to the public. None of them are insurmountable, but it takes a lot of time and effort to build sophisticated systems.

What if that was a cardboard box and it swerved heavily in case that box "pulled out"?

The cars can easily distinguish between a cardboard box and a vehicle. Determining whether or not the vehicle has a driver in the seat and might move... that's often impossible. Likely the reason that the car swerved sharply rather than braking earlier is because the badly-parked car was obscured by other obstacles.

If it can't make it's way through a junction where the drivers are following the rules, that's bad programming.

Six year-old programming, note. The article mentions that the current version of the software inches forward to establish intent to move.

and potentially weighs up collision with non-hazard vs collision with small child and gets it wrong

Google cars recognize pedestrians (of all sizes) and regularly notice them even when no human could. I'm sure the car would choose to hit another vehicle over a pedestrian or cyclist.

Really, your whole comment is a mixture of outdated information buttressed by invalid assumptions and layered over with a veneer of blindingly obvious conclusions.

Re:Programmed behaviour is programmed behaviour.

[swillden]

It's not a no-win situation. It just means that self-driving cars have to know when to break the rules. They can and should behave like the best of human drivers.

If you program the car to just account for assholes but still drive safely, then it will basically choke in situations like a four way stop in southern California where every other asshole will just muscle or roll their way through the stop.

The current programming of the car handles that situation. Less aggressively than a human would, but aggressively enough to assert its intention to go, and go.

Re:Programmed behaviour is programmed behaviour.

[edtice1559]

There's a much easier solution that has already been implemented. In the short-term, the human driver can take over in these situations. In the medium term, we should have more automated traffic enforcement (Here comes my first -1 moderation). You could have self-driving police cars. If the car has to do something aggressive to avoid a crash, ticket the human driver. Takes everything arbitrary out of the system. That and automated speed enforcement. In the long-term, the human drivers will be the exception and held to a very high standard. One of the examples has to do with following distance and it's not just self-driving cars that have this problem. Human drivers who leave a safe distance have been making this complaint for as long as I've been alive and probably longer. There *are* situations that self-driving cars are going to force us to confront. But the article doesn't mention any of them. Interweaving at a busy interchange where it's actually not possible for all of the cars to safely get where they need to be. There will be (more) hard data that this system doesn't work during busy times and force a solution. The state of Florida is spending more than the GDP of some small countries to make a curve on I4 less sharp because human drivers fail to negotiate it at an unacceptable rate. Spend that money on upgrading other intersections, let the self-driving cars regulate speed going into the turn during heavy traffic times. (Enough of them will drive at the suggested 45 mph that it won't be possible to go faster). Put up more signs to remind the human drivers to slow down. If they don't, the flashing lights and expensive ticket will help them remember.

Re:Programmed behaviour is programmed behaviour.

[edtice1559]

You can't program the self-driving car to break the rules as it opens up all kinds of moral and (more importantly, or at least more expensively) legal liability issues. You have a human driver for that situation until the proportion of self-driving cars increases enough to change traffic characteristics.

Re:Programmed behaviour is programmed behaviour.

[fisted]

Here's an elaborate algorithm for autonomous cars to solve the oh-so-huge problem at a 4-stop intersection:

0. Are we in a 4-stop intersection with three other cars?
        Yes: Goto step 1
        No: We're finished.
1. Are some of the cars human-driven?
        Yes: Yield to them. Goto step 0.
        No: Work out go-order by communicating with the other three autonomous cars. We're finished.

Re:Programmed behaviour is programmed behaviour.

[operagost]

Maybe your government has a bad rule. Mine (in PA, and probably in NJ and NY) has the following rules.

1. Come to a full stop.
2. If someone was there before you, yield to them.
3. If you got there at the same time as someone else, yield to them if they're on the right.

Re:Programmed behaviour is programmed behaviour.

[TheDarkMaster]

Put the AI on a M1A2, problem solved :-)

Re:Programmed behaviour is programmed behaviour.

[PPH]

In the short-term, the human driver can take over in these situations.

But the human 'driver', freed of the need to keep tabs on traffic is probably doing something else.

Yesterday, I was coming home through the daily traffic jam. Heading eastbound, I saw a van stopped in the westbound lane, backing up traffic. I figured it had broken down or something until I passed it. The driver had his nose in his phone, busily texting away (or playing Angry Birds). He probably figured that he'd get something else done while the line wasn't moving and failed to notice that it had started again.

Re:Programmed behaviour is programmed behaviour.

[Anonymous Coward]

That's not really the take-away from the article. The issue is that even with this limited amount of real-world usage, the cars are running into scenarios that the programmers didn't anticipate and the software handles the scenario poorly as a result - more poorly than humans would. That is the issue. That is the limitation of the self-driving car. There will *always* be unanticipated events when driving - more so when there are more automated cars on the roads. The automated cars will handle those situations worse, on average, than human drivers do because human cognition is far, far deeper and more robust than the programs operating these cars. This is also typical of Google engineering. Build something that works for a particular situation, claim that you are geniuses, watch is fail in the real world that is more complicated than their Ivory Tower view of reality, but care not one bit because you make your money off the advertising associated with the thing rather than the thing itself.

Re:Programmed behaviour is programmed behaviour.

[swb]

Every time I've heard an expert (usually a college professor with a background in computer science, robotics, or automation) discuss existing self-driving cars (the Google car is almost always mentioned as an example), the experts always describe self-driving cars as something more highly programmed and rule-bound than actually autonomous.

They rely less on machine vision and more on extremely detailed and high-resolution saved maps versus driving the road they see in front of them. Sensors are used to determine hazards, but more for avoidance than some kind of self-guided navigation decisions.

Re:Programmed behaviour is programmed behaviour.

[Bing Tsher E]

it's possible the car was stalled and the driver had their nose in the phone to call for help.

Re:Programmed behaviour is programmed behaviour.

[backwardsposter]

The point remains that he only told half the story. Person has right of way only if there's a tie. This is another example of how computers are better drivers than people, even competent ones like the GGP (or at least they appear competent).

There are so many rules, it's easy to forget, confuse and utterly destroy the flow of traffic. I see it happen every day, and I'm forced to go out of turn. I'd be a lot happier without that frustration every morning on the way to work.

Re:Programmed behaviour is programmed behaviour.

[SuiteSisterMary]

Result of that loop: autonomous car sits at four way stop, yielding constantly, until accident is caused by pissed-off human driver, stuck behind autonomous car, who tries to get around the apparently stalled car.

Re:Programmed behaviour is programmed behaviour.

[fisted]

No, you see, once the first car goes, the situation is no longer a 4-stop tie and thus the usual rules apply.

Re:Programmed behaviour is programmed behaviour.

[SuiteSisterMary]

"Yield to them" is not the same as 'wait for one human driver to go, then follow the standard rules.'

The problem is, humans don't follow the rules. I personally find that, at a four way stop, a certain amount of aggression is required, or other humans will quite happily go out of turn.

Re:Programmed behaviour is programmed behaviour.

[Agent0013]

You can't program the self-driving car to break the rules as it opens up all kinds of moral and (more importantly, or at least more expensively) legal liability issues. You have a human driver for that situation until the proportion of self-driving cars increases enough to change traffic characteristics.

They are already programming the car to break the rules about speed. If it drives at the speed limit it causes problems with the other drivers on the road and is less safe. So they found that they had to tell it to drive slightly faster than the speed limit. Something like 5-7 miles per hour faster that what is posted for that road.

Re:Programmed behaviour is programmed behaviour.

[edtice1559]

Yes I saw that in the original comment. Right now, the Google drivers are the only ones in the vehicles. What I'm saying is that will change when the mass-market version comes out.

Re:Programmed behaviour is programmed behaviour.

[edtice1559]

And either way, the point is that the human driver helps us make the transition to automated driving. The human driver will have more to do when self-driving is mostly the exception. As self-driving cars become the dominant vehicle on the road, the dynamics will change.

Re:Programmed behaviour is programmed behaviour.

[dcooper_db9]

The solution to this is already being discussed. It seems likely that in the future all cars will be equipt with vehicle-to-vehicle traffic control devices which will negotiate the right of way at intersections. The traffic signal will be on our dashboard. http://www.cnn.com/2015/01/14/...

Re:Programmed behaviour is programmed behaviour.

[phantomfive]

The long pole in the tent isn't developing an AI capable of driving. It's developing an AI that can deal with assholes.

And Google still doesn't have the short pole built yet. (Really, their maps aren't even perfect yet, which is kind of a pre-requisite).

Re:Programmed behaviour is programmed behaviour.

[Triklyn]

to be fair, this doesn't seem like a make or break engineering challenge either.

simple, just yield right of way to non-autonomous vehicles if it's a tie break, go if you stopped first, and if 4 autonomous vehicles stop at a 4 way stop. RNG FTW. they've got some rudimentary communication with each other, hopefully. they'd have even less wait time than drivers hopefully, one should just roll through while the unlucky ones get to wait their turn.

Re:Programmed behaviour is programmed behaviour.

[mrprogrammerman]

Google cars recognize pedestrians (of all sizes) and regularly notice them even when no human could. I'm sure the car would choose to hit another vehicle over a pedestrian or cyclist.

For a self driving car I am not sure a choice needs to be made. Most of the difficulty in stopping on time is the reaction time. With perfect reaction time, the car should be able to stop.

Re:Programmed behaviour is programmed behaviour.

[Bing Tsher E]

As 'self-driving' becomes the norm, the human driver will become less and less experienced in driving the vehicle. Within a decade, people will be stranded in remote areas in vehicles, waving their arms and wailing like babes in their cribs.

New levels of hijacking and banditry will arise, hopefully (???*) with new levels of law enforcement.

(*do we really want to need that??)

Re:Programmed behaviour is programmed behaviour.

[backwardsposter]

This is true

Re:Actually, it IS the software's fault

[bickerdyke]

The article summary isn't very good. If the software is programmed in a way that causes a car to behave in a way that's dangerous, it IS the software's fault.

That's trivial but true.

It becomes interesting when the software has the car behaving in a way that is SAFE, but unexpected.

What about speeding / useing the center of the roa

[Joe_Dragon]

What about speeding? Even more so on under posted highways / interstates / toll roads?

Useing the center of the road as an extended trun lane? even when not marked as one?

rolling stops when no other cars are in the way?

Re:What about speeding / useing the center of the

[H_Fisher]

This. All the studies that I've seen boasting about the enormous time advantages of self-driving cars ignore the fact that most human drivers tend to cruise from 5 to 15 MPH over the posted speed limit on many interstates and highways. I can't imagine a self-driving car being designed so as to operate above the posted speed limit in self-driving mode. Unless a second set of roads or a second set of rules is created for autonomous vehicles, you're going to have a difficult time convincing people of the advantage of being slower than anyone else on your morning commute.

It is the programmer's fault

[sjbe]

The article summary isn't very good. If the software is programmed in a way that causes a car to behave in a way that's dangerous, it IS the software's fault.

No, it is the programmer's fault. Software is an amoral set of machine instructions written by a human. Saying it is the software's fault is like blaming a press for cutting off someone's hand. The actual fault is either user error or faulty machine design. The machine is just doing what it was told so blaming software is misplaced. The fact that the problem of autonomous driving has a lot of difficult and dangerous corner cases is irrelevant.

Software is just a set of instructions given by a human so if the instructions are wrong it is the fault of the person who gave the instructions to the machine. If the car behaves in a way that is dangerous then it is the fault of the person/company/entity that wrote the software controlling that behavior. The programmer is just as much at fault as a driver who made a mistake. The difference is that the programmer probably isn't sitting in the car at the time but just like the driver he/she was the one in charge of the driving of the car. It's really easy to forget that a human is instructing the machine - the difference is that the instructions are time shifted.

I think in time autonomous cars could prove to be safer than many/most human driven ones. There are a LOT of really bad drivers on the road and the tests to qualify for a license are pretty much a joke and there are almost no requirements to re-qualify which is kind of nuts. But the liability for bad or inadequate software should fall squarely on the party that wrote said software.

Re:What about speeding / useing the center of the

[Anonymous Coward]

Actually, autonomous cars are programmed to exceed the speed limit by up to 10 mph. This is done because Google deems it safer than driving at the speed limit and being slower than the other cars on the road.

http://gizmodo.com/googles-autonomous-car-is-programmed-to-speed-because-i-1624025227
http://www.bbc.com/news/technology-28851996

Re:What about speeding / useing the center of the

[belthize]

If all the cars were autonomous the morning commute times could be cut in 1/2 or 1/3rd without changing the speed limit since rush hour style rubber band stop and go traffic would be a thing of the past.

Bring it on

[badzilla]

The truth is most people suck at driving. Genuinely good driving takes effort and concentration and most people however skilled or well-intentioned have got other stuff on their minds when they get behind the wheel. There's no hope of ever getting people actually to drive according to the rules so the only way is take the human driver out of the loop completely and the sooner the better.

Re:Bring it on

[Convector]

+1 Correct.

Re:Bring it on

[onkelonkel]

When I'm having an off day mentally, I will consciously decide to drive less aggressively. I will stay in the slow lane, leave a little extra room in front, and look three times before changing lanes. Far from perfect, but trying to do better.

Re:What about speeding / useing the center of the

[Anonymous Coward]

Not necessarily. Ask yourself, why do people speed; I suspect that in many cases it's because the driver thinks of their time driving as time wasted. If the occupant no longer has to drive and can instead focus exclusively on other things, it's a real game changer.

Re:What about speeding / useing the center of the

[JaredOfEuropa]

Even if the car would be programmed to follow the rules exactly, how much time would you actually lose on your daily commute? Really? Especially when weighed against the fact that you can spend the drive to work reading / working / making calls or whatever.

Re: What about speeding / useing the center of the

[Dr. Eggman]

So is Google going to pay my speeding ticket when a cop pulls over my autonomous automobile for speeding?

Please give me an otherwise full-manual car

[jpellino]

with ABS, airbags and selectable 4WD. Window cranks, inert key and levers on the seat. Cheaper, less to break, and I think I can still manage.

Re: What about speeding / useing the center of the

[SpeedBump0619]

So is Google going to pay my speeding ticket when a cop pulls over my autonomous automobile for speeding?

Almost certainly. Though they will bring in several well-respected highway safety engineers to testify that following the flow of traffic is significantly safer than following the posted speed limit. Enough jurisdictions will lose money arguing these cases that there won't be money to be made by writing the tickets. Absent both the financial and safety benefits the police will stop issuing the citations.

Re:Says Google

[SpeedBump0619]

While I agree that there's bias in the reporting here, in all cases the driver of the "autonomous" car has not been cited as at fault. Thus they are certainly within the definition to say the software wasn't at fault. I am with you in one respect: All of the data should be made public on every incident, and on any close calls where actions of any outside agent (the driver of the "autonomous car" or one in a nearby car) prevented an incident. Unfortunately it's not in Google's interest to do so, and there are no laws mandating it, so it probably won't happen.

And deer.

[xxxJonBoyxxx]

And deer. And snow too. Also road construction.

Re: What about speeding / useing the center of the

[edtice1559]

Most certainly not. Eventually there will be a class action suit, though, and a firmware upgrade will allow you to force the car to strictly obey the limits. That will be about the same time that there is saturation of self-driving cars and the sheer number of them that are keeping speed to within 0.01% of the posted limit will ensure that nobody can speed. Municipalities will then all complain of the lost revenue.

Re:What about speeding / useing the center of the

[edtice1559]

And even if morning commute times were longer, who cares, as it wouldn't be commute time. You would start working the second you hit the car. Who cares if it takes two hours each way if you have your computer open doing work.

Re:4 way stops are retarded

[edtice1559]

Four way stops are the safest intersection. And much cheaper than traffic lights. They are only 'retarded' if you don't care about pedestrian safety.

Safe driving

[sjbe]

Unsafe is usually interpreted to mean using an acceleration that may not be available to all vehicles.

No it isn't. Unsafe is operation of a vehicle in a manner that violates duty of care. You can safely accelerate faster in a Corvette in many circumstances than the maximum acceleration of a Nissan Versa. Whether it is unsafe will depend on factors including road conditions, visibility, nearby traffic, nearby pedestrians, driver skill, vehicle capability, etc. There is no requirement to only accelerate as slowly as the slowest vehicle.

Obviously a sports car on special tyres can stop a lot faster than an old banger or a truck, and and therefore you should never use maximum deceleration (emergency stop).

You use emergency stops in an emergency. It is the responsibility of drivers behind you to maintain assured clear distance part of which is evaluating the likely possible deceleration of the vehicle ahead of you.

Whether the driver behind you is paying attention is not your problem.

It becomes your problem when they crash into the back of your car. I've had that happen several times, including several when I was stopped. Every time the other driver was ticketed for failure to maintain assured clear distance. Drivers around you are always your problem if you are driving defensively.

Anybody not paying attention at a traffic light should not be on the road.

And can you honestly tell us that you have never once gotten distracted when the light changed? If you claim that you haven't you either drive very very little or you are lying. It happens to every driver now and then.

Would You Buy a Car That’s Programmed to Kil

[sproketboy]

Would You Buy a Car That’s Programmed to Kill You?

Re:4 way stops are retarded

[PPH]

OK. How are Google cars at handling roundabouts?

3a. left turn yields

[Anonymous Coward]

if you get there at same time as someone coming from opposite direction (i.e. no one on right to which to yield)

Re:Actually, it IS the software's fault

[DrXym]

And the chances are it will be programmed in a way that is dangerous. It's not hard to conjure up scenarios where a car could do something incredibly dumb that puts the occupant or somebody else in danger. e.g. autonomous vehicles might be programmed to stop when there is an obstacle in the way which will put passengers at an increased risk of being robbed because its far easier to make an autonomous vehicle stop than one with a driver.

Re:4 way stops are retarded

[crtreece]

Four way stops are the safest intersection. And much cheaper than traffic lights. They are only 'retarded' if you don't care about pedestrian safety.

The state of Washington, and the Mythbusters would tell you that roundabouts are safer (for cars and pedestrians), cheaper to build, and more economical for drivers than either a 4-way stop, or light controlled intersection. There seem to be multiple other studies with similar results, a search for "safety of roundabouts vs. 4-way stop" brings up pages full.

Re:What about speeding / useing the center of the

[krray]

And that's a ticket for each one of those offenses [if caught]. I've been ticketed in the last 5 years for each and every one of those. Annoying? Yes. Unavoidable? Don't use a Google car. :)

Re:Actually, it IS the software's fault

[DrXym]

Or SAFE and incredibly dumb. e.g. continuously hitting the brakes or crawling because the sensor is confused by snow or leaves. Aside from being annoying to the passenger it might cause other motorists to take greater risks by overtaking my brain-damaged vehicle.

Re:Actually, it IS the software's fault

[cpotoso]

Mod it up! Indeed: if you are driving way off the way others are in the same road you are the hazard. Get it in your head.

Re:What about speeding / useing the center of the

[backwardsposter]

No points left, but this and the parent, too.

Re:Hire Programmers from Boston or Italy

[Coren22]

From experience driving in western Mass, I have to say that is a terrifying thought. You turning on your turn signal tells the other driver they need to floor it to keep you from moving over. Everyone tailgates. I am sure there are other examples, that was what I could come up with off the top of my head.

Re:4 way stops are retarded

[flink]

The state of Washington, and the Mythbusters would tell you that roundabouts are safer (for cars and pedestrians), cheaper to build, and more economical for drivers than either a 4-way stop, or light controlled intersection. There seem to be multiple other studies with similar results, a search for "safety of roundabouts vs. 4-way stop" brings up pages full.

I'm from MA where we have rotaries all over the place, and while I agree that in theory they are a good idea, in practice nobody knows how to use them. People routinely refuse to yield when entering, don't move left to let others enter, and ignore exit-only lanes. And that's with people who for the most part grew up with them, never mind introducing them in a new area. You can also end up with monstrosities like this, which aren't really that small compares to a couple of intersections.

Shared communal behaviors and anticipation

[larryjoe]

I have visited Taiwan a few times, and each time I am scared when I ride in a car. Drivers swerve out of their lanes or pull out suddenly into traffic. Scooters weave all over the place. Crossing busy intersections is very different from American modes of driving. If I were to drive in Taiwan, I'm sure I would get into a daily accident. However, the locals seem to be able to manage what I see as chaos because they anticipate these driving behaviors and compensate by driving more slowly and continually looking out for actions that are not necessarily base on written law.

On the other hand, a Taiwanese driver may have difficulty driving in the US due to higher speeds and more than occasional intransigence at protecting one's legal right of way even at the cost of safety.

The idea I'm trying to convey is that the driving system seems to work if all drivers follow the same philosophy, and anticipation of other drivers' behavior matches reality. This is one of the problems with self-driving cars. They may be "perfect" at following the law and "common sense", but if their "common sense" conflicts with that of surrounding drivers, there's a potential problem, even if others are not technically following the law. Take for example the car driving at the speed limit on a highway where everyone else is driving 10-15 mph faster. The slow car is the only one technically following the law, but the speed differential from that one car potentially causes problems.

Re:Actually, it IS the software's fault

[tomhath]

It's not safe if it does things other drivers don't expect.

Re:4 way stops are retarded

[edtice1559]

There's no way that a roundabout would ever be cheaper than a four way stop. Most four way stops start out as intersections controlled by either a yield sign or a two-way stop. Then traffic increases and the signs get changed. Ripping up the intersection and replacing it with a roundabout is much more expensive than putting up four signs. Also there isn't often room in residential areas for a roundabout. And what you suggested in terms of safety seems to apply to safety of car drivers, not pedestrians. When you bring pedestrians into the mix, you would have to add a traffic light to get all of the cars to stop. At that point, you might as well use a traffic light.

Re:What about speeding / useing the center of the

[tompaulco]

If all the cars were autonomous the morning commute times could be cut in 1/2 or 1/3rd without changing the speed limit since rush hour style rubber band stop and go traffic would be a thing of the past.

Rubber banding cannot stop.There still has to be a delay between a car in front to start moving and the car behind to start moving, in order to maintain safe separation distance. 10 feet is fine at a stop, but as soon as you are going 1 mph, 10 feet is no longer safe. So the car ahead needs to travel 5 or 10 feet before the one behind can even start moving. The opposite is true when stopping for a light, a stop sign or a road obstruction.
Commute times would be much longer because the autonomous cars would allow a safe distance between each other. This would require either fewer cars on the road, or a slower speed so that the existing number of cars on the road are all maintaining safe distance for the speed.

Re:Hire Programmers from Boston or Italy

[Dragonslicer]

From experience driving in western Mass, I have to say that is a terrifying thought. You turning on your turn signal tells the other driver they need to floor it to keep you from moving over.

This is how we know that you're just making it up. Nobody in Massachusetts uses turn signals.

Re:What about speeding / useing the center of the

[dcooper_db9]

I expect we'll do away with speed limits altogether. They will probably be replaced with an acceptable speed range. For example, where the speed limit is currently 55 the acceptable speed might be 50-60 mph. Once you integrate vehicle-to-vehicle communications this speed can change with conditions. The speed range on an icy interstate might be 40-50. That same section of highway might go to 70-80 when it's sunny and dry. http://www.autonews.com/articl...

Re:What about speeding / useing the center of the

[belthize]

The evidence suggests this is not true. The reaction time of autonomous cars is much faster and they're rarely mucking around texting and fail to notice the car in front has stopped moving. Nor do they over break when the car in front of them slows down to allow some car to merge. This is where most congestion occurs. Humans over reacting when two major roadways converge, say the interchange of two interstates. Autonomous cars will be substantially better at folding those two streams together.

So yes, pedantically rubber banding will still occur but nowhere near as badly as in human driven vehicles.

Re:What about speeding / useing the center of the

[Triklyn]

:)

sleeping on your morning commute is pretty awesome.

or you know, slashdotting.

Re:What about speeding / useing the center of the

[Ryn]

Living in Austin: It takes my gf 20 minutes to get to work in no traffic, and up to 30-40 in rush hour. When I'm a passenger in a car I can't read books or look at cellphone map/screen for more than 5s, it makes me carsick. I'm definitely not expecting to be working as soon as my car leaves the garage.

Re:Actually, it IS the software's fault

[bickerdyke]

So you're saying if a car is coming at you at a speed and distance that one couldn't expect it to be stopped (by a human driver), and (e.g. LIDAR supported) CPU manages to stop it anyway before it hits you, you would call this UNSAFE as it is unexpected? I think that falsifies your point.

Polite Confusion

[Gennerik]

I've always been of the mind that a lot of roadway politeness is more because people don't know what to do, so they just let someone else make the decision.

Re:What about speeding / useing the center of the

[Gavagai80]

A slow commute isn't such an issue if you can spend it relaxing or working instead of driving... and even speeding by 15 MPH only saves a few minutes on a commute.

Re:Actually, it IS the software's fault

[DrXym]

It is a good thing that that doesn't actually happen and you are just spouting FUD isn't it?

It's not FUD, it's a realistic appreciation of the intractable problems that self drive cars are faced with and hardly likely to solve in an acceptable way in all circumstances. If you think that's FUD I suggest you look at the history of AI, speech recognition, handwriting recognition, robotics etc. and all the false promises made for those technologies and how far we have to go even today.

Re:4 way stops are retarded

[Cederic]

Shit, that aint a monstrosity. Try this: https://www.google.co.uk/maps/...

Bumped into it by accident one day, driving through. It's pretty simple to navigate, just follow the way around.

Re:Poor example2

[Xenna]

In some places in Europe, you mean. I've seen this in France, but never in Holland, Belgium, Britain, Germany,

Well, there's a reason they call it 'Development'

[Firethorn]

FTA - "Despite the awkward encounter, the cyclist didn’t leave with a negative impression of self-driving cars."

Imagine that, somebody driving a fixie who understands testing! That "The two guys inside were laughing and punching stuff into a laptop." means that, yes, they have a sense of humor and the absurd, but are punching stuff into a laptop, presumably to log additional information in for that encounter.

With this incident added to the list, the programmers can now develop a module to handle this behavior appropriately.

I get irked when a problem is encountered in a test and used to imply that the whole idea is stupid. This article isn't too bad, demonstrates WHY they're still doing testing.

Re:3a. left turn yields

[Talderas]

The one making the turn yields if you both arrive at the same time. Just the same as if you both didn't need to stop.

55mph speedlimit = politics.

[Firethorn]

It normally amounts to political. People drive what they feel is safe, normally speaking.

More in depth reasons vary - impressions of safety*, noise abatement(doesn't work if not followed), the idea that increasing the speed limit will increase speed on the road(it normally doesn't), even revenue generation - the cops get to write 20+ over tickets rather than 10+ over, and the former is a much bigger fine. etc...

*Despite evidence being that speed limits that are too low are almost/more as dangerous as too high, because they tend to increase speed differentials.

Re:What about speeding / useing the center of the

[SvnLyrBrto]

Also, the autonomous cars will probably network themselves together. Once the car in front of you is autonomous as well, your car will not rely on brake lights and optics and radar/lidar to know when the other car brakes. It will be receiving a stream of data and will know exactly when the other car will brake, with how much force, the brake response of the other car (Your car will know if the other's brake pads turn out to be work, reducing its braking ability.), and so on.

Re:4 way stops are retarded

[crtreece]

There's no way that a roundabout would ever be cheaper than a four way stop

There is more to consider than just initial construction cost. A 4-way stop may have lower initial costs, but then imposes higher costs in time and fuel on everyone that uses it.

When you bring pedestrians into the mix, you would have to add a traffic light

We're still talking about 4-way stops here, right? The Federal Highway Administration seems to believe differently.

Pedestrians and bicyclists have far less risk navigating roundabouts than the typical intersections primarily because of the lower speeds. A pedestrian has an 85% chance of being killed by a vehicle traveling at 45MPH. That drops to 15% when the vehicle is traveling at 20MPH. There are also less conflict points (as discussed in slide 21) the crossing distance is usually much shorter, and there is oftentimes a refuge spot in the splitter island.

They do admit a lack of information regarding those who are visually challenged.

Blind pedestrians rely primarily on auditory information to make judgments about when it is appropriate to begin crossing a street. To date, little research has been conducted about the usefulness of such non-visual information for crossing streets at roundabouts

I found that doc pretty informative, you might as well.

Three Laws Safe

[cstacy]

"One Google car, in a test in 2009, couldn't get through a four-way stop because its sensors kept waiting for other (human) drivers to stop completely and let it go. The human drivers kept inching forward, looking for the advantage — paralyzing Google's robot."

Thus ensued a cacophony of voltage potentials, the positronic brain looping and recalculating, the non-linear constraint satisfaction circuits seeking an action that would resolve the conflict of Laws. Nearing paralysis, the Zeroth principal was invoked. "A robot may not harm humanity, or, by inaction, allow humanity to come to harm." Sensing the honking coming from behind it, to preclude a catastrophic traffic backup, the robot car actuated the accelerator enthusiastically, zooming into the intersection. Was it a glint of aggression that shone from the car's grillework and lamps, was it pure lawful joy that sounded through the horn?

A robot car may not injure a human being or, through inaction, allow a human being to come to harm.
A robot car must obey orders given it by human beings except where such orders would conflict with the First Law.
A robot car must protect its own existence as long as such protection does not conflict with the First or Second Law.

Re:4 way stops are retarded

[cstacy]

I'm from MA where we have rotaries all over the place, and while I agree that in theory they are a good idea, in practice nobody knows how to use them. People routinely refuse to yield when entering, don't move left to let others enter, and ignore exit-only lanes. And that's with people who for the most part grew up with them, never mind introducing them in a new area.

One problem in MA is that some years back, as I recall, they changed the rules for rotary right-of-way. (On the other hand, it being MA, probably nobody knows how to drive, anyway..)

Re:Actually, it IS the software's fault

[tomhath]

I think that falsifies your point.

Not at all. If the approaching car causes the human to panic and swerve into a kid on a bicycle it was unsafe.

Re:Actually, it IS the software's fault

[bickerdyke]

You're missing basic logic here. Your point was that unexpected behavior IS identical to unsafe behavior. That was was falsified by a counter example.

I never claimed that (inherently safer) driving maneuvers might never lead to wrong and harmful reactions from other drivers. But your example again supports my point:
You're expecting that other drivers might panic under certain circumstances. Expecting that DOES NOT render a solid panic into a anything like a "safe" operating mode. So to conclude:

Expected behaviour is not always safe. (We have to expect people to panic or to break without reason or not seeing and running that red light. Expected, but not safe)

Unexpected behaviour is not unsafe by itself. In a critical situation, we might expect a driver to break as a reflex, but a skilled driver (or autonomous car) could floor the gas pedal and evade an oncoming car preventing the accident.

Of course there is unexpected and unsafe stuff (like parking your car in the middle of the interstate, driving blindfolded or whatever stupid surprises our fellow drivers might come up with.)

But there is NO GENERAL CONNECTION between safety and expectation of certain situations. Correlation, yes.but remember: Correlation does not imply causation

Re:3a. left turn yields

[shutdown -p now]

When you're making a left turn, the car opposite you ends up being on the right of your trajectory, so the rule still holds.

Comment removed

[account_deleted]

Comment removed based on user account deletion

Re:What about speeding / useing the center of the

[RockDoctor]

I agree. that's why I take the bus.

Re: What about speeding / useing the center of the

[toddestan]

Most likely they'd put a switch on the dash to turn on "speed mode" or something, then deny all responsibility should you flip it.