Slashdot Mirror


Toyota Is Uneasy About the Handoff Between Automated Systems and Drivers (caranddriver.com)

schwit1 shares a report from Car and Driver: Toyota has not yet decided whether it will bring a car to market that is capable of automated driving in some situations yet still requires a human driver behind a wheel who can take control if needed -- but the automaker, characteristically, is more cautious than many about moving forward with the technology. Citing safety concerns regarding the handoff between self-driving technology and human driver, Kiyotaka Ise, Toyota's chief safety technology officer, said the biggest issue with these kinds of systems is that "there is a limbo for several seconds between machine and human" in incidents when a car prompts a human to retake control if it cannot handle operations. These kinds of systems, defined as Level 3 autonomy by SAE, have divided automakers and tech companies in their approaches to developing cars for the self-driving future. As opposed to Level 2 systems, like Tesla Motors' Autopilot, in which a human driver is expected to keep his or her eyes and attention on the road while a system conducts most aspects of the driving, Level 3 is characterized by the system's claiming responsibility for the driving task when it is enabled. Although Toyota assures us that its researchers are hard at work figuring out the challenges of Level 3 autonomy, it seems like the company could eventually join others moving directly from its current Level 2 system to a Level 4 system. Given the self-driving race has been on for a while, this could put Toyota at a competitive disadvantage, but it's clear engineers at the company care more about getting things right than they do about being first.

135 comments

  1. Toyota is... by Anonymous Coward · · Score: 5, Insightful

    The summary states, "it's clear engineers at the company care more about getting things right than they do about being first."

    So, basically what you're saying is, Toyota is the anti-Tesla.

    1. Re:Toyota is... by AmiMoJo · · Score: 3, Insightful

      I'm not convinced that Tesla will get to level 5 with their current hardware. They are already selling level 5 to customers as a future firmware update (â3000 extra last time I looked), saying they will upgrade hardware if necessary for people who already paid.

      Their system is only cameras and ultrasonic sensors, no lidar. They are using neural nets for image processing.

      --
      const int one = 65536; (Silvermoon, Texture.cs)
      SJW, n: "Someone I don't like, and by the way I'm a fuckwit" - AC
    2. Re:Toyota is... by Anonymous Coward · · Score: 5, Funny

      Tesla is the industry leader in making promises .

    3. Re:Toyota is... by Anonymous Coward · · Score: 0

      The summary states, "it's clear engineers at the company care more about getting things right than they do about being first."

      So, basically what you're saying is, Toyota is the anti-Tesla.

      People with more money than brains who fail to RTFM and assume that "autopilot" means "sit-back-and-enjoy-a-movie" are anti-intelligent. Basically what I'm saying is put blame where it belongs.*

      (* = No, I don't own a Tesla or a Toyota.)

    4. Re:Toyota is... by Anonymous Coward · · Score: 1

      To be fair, Tesla constantly presents their mediocre driving aid as if it were a level 5 autonomy system. I can imagine that someone less interested in automotive technology may think that it is.

    5. Re:Toyota is... by mjwx · · Score: 4, Insightful

      The summary states, "it's clear engineers at the company care more about getting things right than they do about being first."

      So, basically what you're saying is, Toyota is the anti-Tesla.

      Basically you're saying Toyota is being Toyota (conservative to the extreme, but good at what they do).

      Toyota is not the only one concerned with this, as a road user I'm concerned what will happen when Dopey Doris' automated car struggles with faded lines on a single lane road (quite common on my 18 mile commute to work). Right now, Dopey Doris can only spend half her attention on her phone, I hate to think what will happen when she puts her full attention into it and because she's so engrossed in FaceCrush or watching the latest episode of Keeping up the Cardasians that she completely tunes out the alarm throwing control back to the driver and the car veers into my lane uncontrolled.

      Autonomous cars need to be 99.999999999999999% reliable before they should be considered ready for public consumption. Right now, they're nowhere near it. Google's success has been due to two factors, 1. it was all done in sunny California (I'd like to see the same car in Berkshire) and; 2. the car has been in the hands of a professional driver the whole time. The current track record for autonomous cars is nil, the record is for car and driver working together. Of course we know in the real world if you gave the Google autonomous car to Dopey Doris commuting from Finchamstead every day, she's going to assume that it will do everything for her, so we need to make sure it can operate without human intervention because it needs to, human intervention cant be counted upon from the average steering wheel attendant with a phone shoved up their nose (we get enough collisions from these types as it is without giving them a reassurance).

      --
      Calling someone a "hater" only means you can not rationally rebut their argument.
    6. Re:Toyota is... by zifn4b · · Score: 1

      The summary states, "it's clear engineers at the company care more about getting things right than they do about being first."

      So, basically what you're saying is, Toyota is the anti-Tesla.

      No, Toyota is being responsible instead of going for a quick short-term money grab due to hype.

      --
      We'll make great pets
    7. Re:Toyota is... by Anonymous Coward · · Score: 0

      Which is exactly what GP said.

    8. Re:Toyota is... by rmdingler · · Score: 2

      Autonomous cars need to be 99.999999999999999% reliable before they should be considered ready for public consumption.

      Since that's essentially an impossible standard, what you're saying is that we should never use autonomous land vehicles, even though human drivers fall massively short of that same safety threshold.

      --
      Happiness in intelligent people is the rarest thing I know.

      Ernest Hemingway

    9. Re:Toyota is... by PolygamousRanchKid+ · · Score: 4, Interesting

      The Duke of Wellington claimed he won the Battle of Waterloo against Napoleon on "the playing fields of Eton".

      The big battle for autonomous driving will be won or lost in the tort courts of the US. Who is responsible for the accident? The driver? Or the manufacturer?

      Your local ambulance chaser lawyer would prefer to sue the manufacturer . . . simply because the manufacturer has more money!

      The first big cases will unsettle the industry, but a sort of fudge agreement will be reached between lawyer groups, the manufacturers and the insurance companies. Unfortunately, the average driver will end up paying for this.

      The lawyers don't want to kill the autonomous car industry . . . they want to "milk" it for their "piece of the action".

      --
      Schroedinger's Brexit: The UK is both in and out of the EU at the same time!
    10. Re:Toyota is... by CastrTroy · · Score: 4, Insightful

      This is also a big concern of mine. Cars should either be 100% autonomous, or 0% autonomous. I'm all for adaptive cruise control, but as soon as you introduce technology that allows people to take their attention away from the driving and have it still follow the road for a significant period of time, that's where you run into problems.

      If you haven't had to actually touch the steering wheel for a month, how much would you really be paying attention? What happens when the car screws up and you need to take over? Are you going to be too engrossed if your other activities to take over? Also, what is the point of paying for all this technology anyway if you don't get to actually not pay attention anyway. If you're going to have to keep your hands on the wheel, you might as well actually be driving, because other wise it isn't really worth the expense.

      --

      Anthropic principle: We see the universe the way it is because if it were different we would not be here to see it.
    11. Re:Toyota is... by AvitarX · · Score: 1

      I actually think Toyota is wrong here.

      I suspect that advanced level 2 (what we currently have) is the most dangerous. An aggressive alarm with a seconds long handover seems far safer than completely unatentive touching the wheel.

      I think the danger zone starts as soon as you combine adaptive cruise control and lane assist, and that the only safe option is to leave the wheel completely in control of the human until level 3 (which I think is safer than current level 2).

      --
      Wow, sent an e-mail as suggested when clicking on "use classic" banner, and got a fast response that addressed my msg
    12. Re:Toyota is... by rmdingler · · Score: 1

      You raise an important point, but as soon as autonomous land vehicles exceed the miles driven per accident that humans are capable of, insurance companies will line up in favor of paying for fewer wrecks.

      Not to put too fine a point on it, but the average driver already pays for this under the mandatory auto insurance laws.

      --
      Happiness in intelligent people is the rarest thing I know.

      Ernest Hemingway

    13. Re:Toyota is... by Anonymous Coward · · Score: 0

      IHS has indicated that adaptive cruise control combined with lane assist is showing a measurable increase in safety.

    14. Re:Toyota is... by Ost99 · · Score: 1

      No lidar, but they have a forward facing radar.

      8 cameras, 1 radar and 8 ultrasound sensors. Way more than a we have.

      --
      ---- Sig. gone.
    15. Re:Toyota is... by bws111 · · Score: 1

      IHS has indicated that lane departure WARNINGS increase safety. That is far different than 'the car is driving until something goes wrong'.

    16. Re:Toyota is... by Anonymous Coward · · Score: 0

      Human drivers should be 99.999999999999999% reliable before they should be considered ready for public roads.

    17. Re:Toyota is... by fluffernutter · · Score: 1

      I"d say at the least the onus is on them to guarantee that it will never create an accident in a situation that I, as an individual, would be able to deal with. And that should go for anyone buying an automated vehicle.

      --
      Laws are rules for the court, but merely a bottom bar to hit for life. Think beyond laws in your actions always.
    18. Re:Toyota is... by AvitarX · · Score: 1

      Maybe I used the wrong word, lane departure warnings are only a plus if they don't false positive an annoying amount and don't turn the wheel so you can't rely on them to make micro adjustments.

      --
      Wow, sent an e-mail as suggested when clicking on "use classic" banner, and got a fast response that addressed my msg
    19. Re:Toyota is... by fluffernutter · · Score: 1

      Some humans that may decide to use an automated vehicle are. That's all that matters.

      --
      Laws are rules for the court, but merely a bottom bar to hit for life. Think beyond laws in your actions always.
    20. Re:Toyota is... by Anonymous Coward · · Score: 0

      If they're better than human drivers, but one of the imperfections happens to affect you, it's unacceptable? Fucking idiot.

    21. Re:Toyota is... by fluffernutter · · Score: 1

      Sure that's unacceptable. Why would I want a technology that will get me into an accident? Why would anyone want that?

      --
      Laws are rules for the court, but merely a bottom bar to hit for life. Think beyond laws in your actions always.
    22. Re:Toyota is... by AmiMoJo · · Score: 2

      The whole sensor package seems poorly thought out. For example, the lack of a nose camera means that they can't implement a 360 degree overhead view like Nissan and several other manufacturers have now.

      And then there is the whole "AP2.5" debacle, where they did a major computing power upgrade over the original AP2.0 hardware after realizing it wasn't up to the task, and then promised free upgrades to people who had already paid for full self driving a year ago.

      The thing about cameras is that they are not the same as human eyes. People have been extracting images from the Tesla cameras and the resolution just isn't good enough to resolve signs at distance, for example. Also processing those images is proving to be difficult. Currently they are down-sampling them to an even lower resolution before handing them to their neural networks for recognition.

      The current state of their system is that they can follow well marked roads somewhat reliably. On corners the car will sometimes stray into oncoming traffic or into walls. Junctions confuse it. Even on well marked roads in good weather it sometimes ping-pongs left and right in the lane. And they use GPS maps for assistance, something that a full self driving car can't rely on.

      Musk said they would demo driving from coast to coast autonomously this year. Seems unlikely to happen, although it might just be possible if they stick to major roads. But they are so far from being able to handle anything tricky, like roundabouts or unusual junctions or car parks or even read basic road signs and traffic lights that I'm thinking some of the people who already paid for the self-driving feature will be selling the car long before it ships.

      --
      const int one = 65536; (Silvermoon, Texture.cs)
      SJW, n: "Someone I don't like, and by the way I'm a fuckwit" - AC
    23. Re:Toyota is... by Rei · · Score: 2

      12 ultrasonic sensors on Model 3.

      That said, what we have that a car doesn't is a brain that automatically fixes photogrammetric stitching errors based on the logic of "that doesn't make sense" - whether something "makes sense" or not being an AI-hard problem. So one tries to compensate for this on vehicles with better sensors than human beings have.

      Lidar provides a superb data stream when conditions are right, but it's too problematic. You're not going to put awkward, draggy, ugly domes on top of everyone's cars. They're way too expensive, and even if the price can come down, they're way too prone to adverse weather. Wherein, what, do you tell people that the car can't drive? Or that they have to?

      I think the future on this front is time-of-flight cameras as a supplemental datastream to multi-input driving systems. A company like Tesla could simply swap out their existing cameras for time-of-flight ones and get a LIDAR-like datastream, which they can plug in as a better alternative to their current photogrammetry + radar streams when weather allows for it, and otherwise fall back on photogrammetry + radar. Time-of-flight cameras still yield the same sort of visual information as regular cameras, but can also record the length of time between a IR pulse is issued and when different parts of the CMOS/CCD register the reflection, and thus produce a high resolution "LIDAR" map instantaneously with each pulse throughout the camera's entire field of view. Because they're manufactured with traditional semiconductor methods, in theory they should be little more expensive to produce when in mass production than conventional cameras.

      Either way, though, you still need conventional cameras for object identification, for seeing colours, and a whole host of other things. A z depth map on its own isn't enough.

      --
      The most exciting phrase to hear in science, the one that heralds new discoveries, is not âEureka!â(TM), but
    24. Re:Toyota is... by Rei · · Score: 0

      To be actually fair, Tesla continually does precisely the opposite, and there is essentially zero confusion among actual Tesla owners about what it's actually capable of. Heck, if you take your hands off the wheel for too long too many times in a row, the vehicle will revoke your autopilot privileges until the next time you charge up.

      EAP is in a state at this point where its limitations are obvious enough that it's hard to become too complacent. The problem arises when autopilot-style systems get good enough that you don't see its flaws very often, and so drivers become complacent - and then when something actually does go wrong, they're not ready to deal with it.

      --
      The most exciting phrase to hear in science, the one that heralds new discoveries, is not âEureka!â(TM), but
    25. Re:Toyota is... by Anonymous Coward · · Score: 0

      No. They're not. And what you've just said is fucking idiotic on many, many levels. I hope you get killed by a drunk driver, because the only way for your particular brand of stupidity to not be completely toxic to our species is for it to serve as part of an ironic warning.

    26. Re:Toyota is... by Anonymous Coward · · Score: 0

      This might shock you, but nobody other than your mother is going to go along with it while you pretend you're perfect.

    27. Re:Toyota is... by fluffernutter · · Score: 1

      Where did I say drunk drivers shouldn't use this technology? All I said was that I don't want to use technology that may get me into an accident, which I think is a perfectly reasonable stance. Especially until the day that insurance is paid for entirely by the technology company in question and not myself.

      --
      Laws are rules for the court, but merely a bottom bar to hit for life. Think beyond laws in your actions always.
    28. Re:Toyota is... by aaarrrgggh · · Score: 1

      I live in Los Angeles and don't own a car, but I have to rent one to drive ~100-250 miles per day a few times per month. Last week I was twice stuck driving at odd hours and struggling to stay awake and focused on the freeway. While I "powered through" it, having just automatic lane keeping and car following operational would have made the trip much less stressful, and likely improve safety by reducing event risk to ~5% per 100 miles to ~0.01%, on par with a non-fatigued driver.

      So then you get into the question of it causing an accident. The most likely scenario is the car deciding to stop and be rear-ended, assuming it has accurate positional/rate of change information on its surroundings. I have to think that it is no worse to a human driver in that regard, at least with dome lidar.

    29. Re:Toyota is... by dehachel12 · · Score: 1

      The problem arises when humans get good enough that you don't see its flaws very often, and so drivers become complacent - and then when something actually does go wrong, they're not ready to deal with it.

      FTFY.
      There were 1.25 million road traffic deaths globally in 2013 that's 400 9/11's.

    30. Re:Toyota is... by fluffernutter · · Score: 1

      Even if I'm not a perfect driver, statistically there are people that come closer to it than I do. Self driving needs to be better than them.

      --
      Laws are rules for the court, but merely a bottom bar to hit for life. Think beyond laws in your actions always.
    31. Re:Toyota is... by Anonymous Coward · · Score: 0

      Tesla have made a few of these videos where they show off how their Autopilot system drives a Tesla car for a bit through a place where pedestrians and cyclists don't exist in perfect weather with no visible driver interventions. Now, to someone with basic knowledge of automobiles and artificial intelligence, this may not seem like an especially impressive feat, but to the kind of person who would buy a Tesla, it may seem as though Autopilot is a high-level autonomy system after seeing such a video. There have been several incidents where Tesla drivers engaged in other activities for extended periods, under the impression that the Autopilot would deal with any and all circumstances that may occur, only to crash into a an obstacle that a human driver or a more advanced driving assistance system would have noticed and avoided very easily.

    32. Re: Toyota is... by Anonymous Coward · · Score: 0

      No, that's almost 1 527 778 9/11s.

    33. Re:Toyota is... by TheCastro1689 · · Score: 2

      Currently, there are about 218 million drivers and in 2016 there were 6,296,000 police-reported motor vehicle traffic crashes. So that's about a 2.8881% crash rate. Just saying.

    34. Re:Toyota is... by Anonymous Coward · · Score: 0

      > Autonomous cars need to be 99.999999999999999%

      You just killed 1 million people/year. Are you ready to take responsibility of your actions? If we have for example 98% working automated cars, they are still much better than humans and can save more than 1 million lives per year. Every day we don't take automated cars into use will cost lives. So decision for not taking them into use should take the current system into consideration.

    35. Re:Toyota is... by Rei · · Score: 1

      Ahem.

      I can only guess that those other 2% have never actually used the thing or known anyone who has.

      --
      The most exciting phrase to hear in science, the one that heralds new discoveries, is not âEureka!â(TM), but
    36. Re: Toyota is... by Anonymous Coward · · Score: 0

      2% is enough to cause quite a few accidents, even if the 98% that claim to understand the limitations act accordingly. Note also that this is the outcome of a study commissioned by Tesla to defend its use of a misleading name.

    37. Re:Toyota is... by Immerman · · Score: 2

      Why? If it's safer than 90% of drivers, then that 90% would be better off in an autonomous vehicle. The remaining 10% would be better off driving themselves, but even they'd benefit from the rest of the traffic being autonomous - fewer idiots to deal with.

      --
      --- Most topics have many sides worth arguing, allow me to take one opposite you.
    38. Re:Toyota is... by Immerman · · Score: 1

      I mostly agree. I could see there being room for a car that safely pulls over, or otherwise gives the driver plenty of time to "shift gears" and develop situational awareness before they have to take over in unusual circumstances - but it should *never* require the driver to take over on short notice, which demands that they be constantly maintaining situational awareness against the very low chance that they need to take over. Human attention just doesn't work that way.

      Highway driving is probably a good example - if the car can handle anything it might encounter on the highway - including cross-parked semis and other "shouldn't be there" problem situations, then you've got an incredibly useful tool. It doesn't need to be able to handle the complexity of city streets, it just needs to be reliably capable of getting the driver safely from on-ramp to exit ramp and hand over control in a safe manner, even if the driver has fallen asleep in the meantime. If that means a string of cars parked along the shoulder a mile before the exit-ramps, as they try to wake their particularly deep-sleeping drivers, I don't see a problem.

      --
      --- Most topics have many sides worth arguing, allow me to take one opposite you.
    39. Re:Toyota is... by apoc.famine · · Score: 1

      Autonomous cars need to be 99.999999999999999% reliable before they should be considered ready for public consumption.

      No, they really do not need to be that reliable. Not even close to that reliable.

      People are not that reliable. People like the Dopey Doris you describe can't keep themselves alive because the phone is so much more interesting than driving is. I absolutely want Doris to have a self-driving car, because even current technology is likely better at driving than she is.

      Even a shitty self-driving car will likely be better than the bottom 25% of human drivers. Unless that makes the middle 50% far worse, it will be a net win. I get the concern, but I think it's 5-10 years out of date. If Doris' car can keep itself in between the lines and auto-brake, that's a win for everyone on the roads.

      Yes, your road may need to be repainted, and it may take Doris crashing into you to drive that improvement. Maybe it's not up to spec, and the entire road needs to be rebuilt. But once that happens, all the rest of the Dorises will not be crashing into people. That's progress. And it doesn't take 100% reliability to get there.

      --
      Velociraptor = Distiraptor / Timeraptor
    40. Re:Toyota is... by fluffernutter · · Score: 1

      Ok well I trust that 90% to know who they are. I'm fine with the risk I incur on myself by driving manually, so I'm not in the demographic for self driving cars.

      --
      Laws are rules for the court, but merely a bottom bar to hit for life. Think beyond laws in your actions always.
    41. Re:Toyota is... by fluffernutter · · Score: 1

      The sensor package doesn't really matter until someone figures out how to make it work under a layer of ice. People are generally good at brushing off their cars but no one will spend time picking ice off the sensors. Maybe current sensors won't be affected by this, but I know cameras will and if there aren't enough working sensors in the most dangerous time of year it's a big problem.

      --
      Laws are rules for the court, but merely a bottom bar to hit for life. Think beyond laws in your actions always.
    42. Re:Toyota is... by fluffernutter · · Score: 1

      Hey if the 10% can point the finger at the car company when they get in an accident and have no financial responsibility or injury then fine.

      --
      Laws are rules for the court, but merely a bottom bar to hit for life. Think beyond laws in your actions always.
    43. Re:Toyota is... by zifn4b · · Score: 1

      Which is exactly what GP said.

      Can you read? The GP said:

      So, basically what you're saying is, Toyota is the anti-Tesla.

      What I said is not logically equivalent to that in any universe.

      --
      We'll make great pets
    44. Re:Toyota is... by Immerman · · Score: 1

      Last I heard the auto manufacturers were pretty unanimous that they would be responsible for any accidents that occur while the car is in self-driving mode, so there's not really a whole lot of conflict to be resolved. It's a simple product-safety liability situation - if you're using a product in full accordance with the manufacturer's instructions, and it injures or kills you anyway, then it's a pretty open-and-shut case of liability for a faulty product.

      The only "loophole" I've seen so far is for "semi-autonomous" vehicles such as the Teslas, which require you to do nothing, but still pay constant attention to the road just in case the car screws up or encounters a situation it doesn't know how to handle. Human attention doesn't work that way, and so such requirements should really be declared unreasonable and void. Presumably most/all existing semi-autonomous vehicles would have the capability disabled, and class-action lawsuits would sort things out with any customers who are unhappy with whatever blanket compensation is (not) offered.

      Alternatively I suppose insurance companies will take up the slack, and semi-autonomous vehicle owners will have to pay an adjusted rate roughly in line with the risk profile. Frankly, as long as the autonomous mode is demonstrably not notably more dangerous than a human driver that seems like a sufficient short-term "patch"

      --
      --- Most topics have many sides worth arguing, allow me to take one opposite you.
    45. Re:Toyota is... by fluffernutter · · Score: 0

      Ok as long as people don't have to pay for it with personal injuries or financial loss.

      --
      Laws are rules for the court, but merely a bottom bar to hit for life. Think beyond laws in your actions always.
    46. Re:Toyota is... by Anonymous Coward · · Score: 0

      How is it different?

    47. Re:Toyota is... by AmiMoJo · · Score: 1

      Level 3 is okay. That's where the car can drive itself under certain limited conditions (e.g. on a highway, but not on urban roads) and gives the driver plenty of warning when they need to take over. By "plenty" I mean 60+ seconds, and if you don't take over nothing terrible happens.

      --
      const int one = 65536; (Silvermoon, Texture.cs)
      SJW, n: "Someone I don't like, and by the way I'm a fuckwit" - AC
    48. Re:Toyota is... by religionofpeas · · Score: 1

      Ok as long as people don't have to pay for it with personal injuries or financial loss.

      They will have to pay. But it will be less than they're paying now.

    49. Re:Toyota is... by Anonymous Coward · · Score: 0

      Currently, there are about 218 million drivers

      Probably quite a bit more. There are well over a billion cars in the world.

    50. Re:Toyota is... by swillden · · Score: 1

      I"d say at the least the onus is on them to guarantee that it will never create an accident in a situation that I, as an individual, would be able to deal with. And that should go for anyone buying an automated vehicle.

      What about all of the situations in which you would have an accident, but the automated system would avoid it? Suppose that there are a thousand of those for every one where you'd succeed and the system would fail. That would fail your requirement as stated. Do you really think that makes sense?

      --
      Note to ACs: I usually delete AC replies without reading them. If you want to talk to me, log in.
    51. Re:Toyota is... by fahrbot-bot · · Score: 1

      The summary states, "it's clear engineers at the company care more about getting things right than they do about being first."

      So, basically what you're saying is, Toyota is the anti-Tesla.

      Perhaps what's being said is that Toyota has had issues in the past with software and is trying to be careful about that.

      See: Toyota's killer firmware: Bad design and its consequences

      Barr's ultimate conclusions were that:

      • Toyota’s electronic throttle control system (ETCS) source code is of unreasonable quality.
      • Toyota’s source code is defective and contains bugs, including bugs that can cause unintended acceleration (UA).
      • Code-quality metrics predict presence of additional bugs.
      • Toyota’s fail safes are defective and inadequate (referring to them as a “house of cards” safety architecture).
      • Misbehaviors of Toyota’s ETCS are a cause of UA.
      --
      It must have been something you assimilated. . . .
    52. Re:Toyota is... by fahrbot-bot · · Score: 1

      Autonomous cars need to be 99.999999999999999% reliable before they should be considered ready for public consumption.

      Noting that nothing else is that reliable. Not people, not Verizon, not even condoms - but use the last one with the first two anyway.

      --
      It must have been something you assimilated. . . .
    53. Re:Toyota is... by fluffernutter · · Score: 1

      They will pay less on average perhaps, but some people who were unlucky enough to have their car confuse a truck for a bridge will end up paying more than they otherwise would and that is just plain wrong. Their fault if they trust the technology I guess.

      --
      Laws are rules for the court, but merely a bottom bar to hit for life. Think beyond laws in your actions always.
    54. Re:Toyota is... by Immerman · · Score: 1

      The manufacturers all seem on board with that scenario - assuming the car is actually in autonomous mode and is at fault for the accident. Pretty straightforward product failure scenario in that case.

      --
      --- Most topics have many sides worth arguing, allow me to take one opposite you.
    55. Re:Toyota is... by Anonymous Coward · · Score: 0

      "Last I heard the auto manufacturers were pretty unanimous that they would be responsible for any accidents that occur while the car is in self-driving mode"

      That's easy. They just hand control back to the driver right before the crash, and blame 'driver error'.

    56. Re:Toyota is... by Anonymous Coward · · Score: 0

      This is also a big concern of mine. Cars should either be 100% autonomous, or 0% autonomous.

      100% takes time and ... um, sacrificial beta testers.

    57. Re:Toyota is... by Anonymous Coward · · Score: 0

      They'll clear ice off their sensors if they have to. The car can refuse to go anywhere if its sensors are obscured.

      People don't clear ice off their windows for fun, they do it because they need to see out of them.

    58. Re: Toyota is... by Anonymous Coward · · Score: 0

      Lame response. People should not have to die for technology.
      There are drivers in the world who have perfect driving records. Why can't an autonomous vehicle be 100%?

    59. Re:Toyota is... by Anonymous Coward · · Score: 0

      To be actually fair, Tesla continually does precisely the opposite, and there is essentially zero confusion among actual Tesla owners about what it's actually capable of. Heck, if you take your hands off the wheel for too long too many times in a row, the vehicle will revoke your autopilot privileges until the next time you charge up.

      EAP is in a state at this point where its limitations are obvious enough that it's hard to become too complacent. The problem arises when autopilot-style systems get good enough that you don't see its flaws very often, and so drivers become complacent - and then when something actually does go wrong, they're not ready to deal with it.

      Lets rewind to 2015;

      Tesla CEO Elon Musk drops his prediction of full autonomous driving from 3 years to just 2

      https://electrek.co/2015/12/21/tesla-ceo-elon-musk-drops-prediction-full-autonomous-driving-from-3-years-to-2/

    60. Re:Toyota is... by Anonymous Coward · · Score: 0

      Your understanding of the issue is completely inverted. When you were simply following the lane (90% of what you do as a driver) you were also prepared to react to a changing situation. You had a sense of other cars around you, you had a sense of your safety zones, the feeling of the throttle, the feeling of the tires on the road. Without that information, you would not be able to react, to identify if something was a hazard or was a predicted event - is that car crossing the lane because it is out of control or because they wanted to change lanes?

      If you take out that state of readiness and let the car drive itself at those times, you will not be able to react with the same accuracy. The computer is most likely to fail at those exact moments when the human driver should be the most prepared to react, and the current strategy is to INTENTIONALLY REDIRECT THE DRIVER'S ATTENTION AWAY FROM DRIVING AS OFTEN AS POSSIBLE. Because people don't like the stress of driving.

      The result, of course, will be a decline in driving skills, an increase in accidents, and an increasingly authoritarian enforcement of driving regulation in favor of technology.

      I say we simply skip the semi-autonomous cars part and go straight for the authoritarian utopia: all drivers must be professional certified, cannot be on any medications or have consumed any alcohol prior to driving, driving routes must be planned in advance, and all speed limits will be strictly enforced.

    61. Re:Toyota is... by superdave80 · · Score: 1

      So, auto pilot won't actually automatically pilot your car for you? Then it probably shouldn't be called auto pilot. But Musk just wants to be cool and be able to say that he has the first auto pilot* car.

      *Warning. An 'auto pilot' car won't actually be able to reliably automatically pilot your car. Please leave your hands on the wheel and look forward at all times... just like when you drive the car.

    62. Re:Toyota is... by superdave80 · · Score: 1

      Ahem, that was a self-serving 'poll' setup by Tesla so they could say, 'See, everybody knows that auto-pilot doesn't really MEAN autopilot!". For Gods sake, they even used Musk's stupid comparison to airplane autopilot systems in one of the questions! Yeah, Tesla didn't have any say in preparing those questions! Also, their own numbers don't add up. 99% say that they know that autopilot doesn't really mean fully automatic... but then 7% say that they do think that it is fully automatic?

    63. Re:Toyota is... by vtcodger · · Score: 1

      "I'm not convinced that Tesla will get to level 5 with their current hardware."

      I think your analysis is perfectly valid. One thing though. Human drivers basically have two (at most) pretty good optical sensors with no useful range capability (human stereo vision only works out to about 6-7 meters). The sensors can scan right to left through about 180 degrees and up/down through 30-45 degrees. They are augmented by at most three very limited mirrors and maybe on modern cars by a flaky, small screen rear camera image when the vehicle is actually moving backwards

      Yet on good days, some humans can do a very credible job of observing and interpreting their environment with that crummy sensor set.

      Maybe automated driving really is mostly a software issue. Not that I believe for one second that Elon Musk can cause the necessary software to be written.

      --
      You can't see ANYTHING from a car, You've got to get out of the goddamned contraption and walk...Edward Abbey
    64. Re:Toyota is... by Anonymous Coward · · Score: 0

      Ok, I'm glad you've got it down to that model, but how about some others? how many Tauruses, how many F150s, how many Priuses?

    65. Re:Toyota is... by Anonymous Coward · · Score: 0

      I realize this is pedantic, but the battle of waterloo was definitely not won by Eton gentlemen, so the analogy sucks. Besides only one in eight units in Wellington's army actually being English, Wellington had a wellknown habit of firing highborn English gentlemen commanders in the Peninsular war, and replacing them with lower born but better, often foreign, soldiers. Constant Rebecque, Dutch chief of staff at Waterloo, was drectly appointed by Wellington to the rank of major in Spain in this way.

      This is often remarked on in Dutch and German sources, because one notable disadvantage in Waterloo was that some of the best, and highest commanding English officers, even Hill, did not speak proper French, and couldn't directly communicate with Dutch or German officers. Eton gentlemen (and Wellington) did speak French.

    66. Re:Toyota is... by antdude · · Score: 1

      But nothing is 100%. Not even 99.99...%. :(

      --
      Ant(Dude) @ Quality Foraged Links (AQFL.net) & The Ant Farm (antfarm.ma.cx / antfarm.home.dhs.org).
    67. Re:Toyota is... by Anonymous Coward · · Score: 0

      While I "powered through" it, having just automatic lane keeping and car following operational would have made the trip much less stressful, and likely improve safety by reducing event risk to ~5% per 100 miles to ~0.01%, on par with a non-fatigued driver.

      Someone with half a brain would realize that having lane keeping and adaptive cruise control ("car following") would have just allowed you to focus less on your driving and enter a more relaxed state, and you very likely would have dozed off completely, leaving your car driving by itself and you 100% unable to take over fast enough should the car have encountered a problem that it could not handle on its own.

      Then again, maybe that would be a good thing. It would result in one less idiot on the road, making driving less dangerous for the rest of us.

    68. Re:Toyota is... by Anonymous Coward · · Score: 0

      No, 1 million retarded, irresponsible fucks killed 1 million people/year.

    69. Re:Toyota is... by Immerman · · Score: 1

      If it attempts to hand off control in a situation where it cannot safely do so, it's still a product safety issue.

      --
      --- Most topics have many sides worth arguing, allow me to take one opposite you.
    70. Re:Toyota is... by CastrTroy · · Score: 1

      By 100%, I mean no steering wheel or the system is good enough that you can sit in the back seat if you so desire. This concept of "good enough but you might have to unexpectedly take over once in a while" isn't really that great of an idea because people simply won't be paying attention if they aren't required to pay attention all the time.

      --

      Anthropic principle: We see the universe the way it is because if it were different we would not be here to see it.
  2. A computer is needed to slam down the accelerator by Anonymous Coward · · Score: 0

    and hold it there at random intervals. Humans can only sometimes accomplish this by themselves.

  3. Re:A computer is needed to slam down the accelerat by Rei · · Score: 1

    I remember way back when that there was a site that plugged itself as a "Toyota Simulator" (seems to be down now). When you went to the site, it was nothing but a looping video of a car driving way too fast down a road and a person screaming. ;)

    --
    The most exciting phrase to hear in science, the one that heralds new discoveries, is not âEureka!â(TM), but
  4. I really don't understand the interest here by DeplorableCodeMonkey · · Score: 2

    Slip ups on the road can become fatal in seconds because of the speeds and forces involved. You know people are going to rely on these systems precisely when they should be off the road period or be paying attention. And what happens when a deer decides to bolt out from the woods in front of your vehicle? Are you going to trust that the car can detect a deer?

    This seems like a solution to people who hate the idea of mass transit and transporting goods by trains. Self-driving cars and trucks and hyperloops! FFS, just hire Disney's engineers and building a fucking monorail in most cities and connect them to the suburbs. That would be more than sufficient to raise the quality of life on transit.

    1. Re:I really don't understand the interest here by zifn4b · · Score: 1

      This seems like a solution to people who hate the idea of mass transit and transporting goods by trains. Self-driving cars and trucks and hyperloops!

      I really hope this is not the primary driver behind this technology. If this is true, the idea is to eliminate the cost of CDL drivers and making the most dangerous vehicles on the road automatically driven. Why couldn't they just have a series of underground tunnels specifically for transporting commercial goods where all the automatic transport vehicles could be driven? That way, if an "error" occurs, it doesn't injure anyone? It's probably because it costs too much to do that I'm guessing... (facepalm)

      --
      We'll make great pets
    2. Re:I really don't understand the interest here by michiganbob · · Score: 2

      And what happens when a deer decides to bolt out from the woods in front of your vehicle? Are you going to trust that the car can detect a deer?

      Actually, yes. I trust a computer system that can handle thousands of computations every second much more than I trust a startled, panicky human in that scenario. I don't know if you've ever hit a deer with your car, but they don't exactly give you much warning, regardless of how much you're paying attention to the road.

      Source: Lives in Pennsylvania

    3. Re:I really don't understand the interest here by Kiuas · · Score: 1

      And what happens when a deer decides to bolt out from the woods in front of your vehicle? Are you going to trust that the car can detect a deer?

      Am I going to trust that a computerized system is capable of detecting an object in front of the car an apply brakes faster than a human being?

      Hell yes I am. The human reaction time is around a second (often considerably more if the driver's distracted or tired). By the time your brain has gone 'oh shit a deer", and decided to slam the breaks, the computer is already in the process of breaking or adjusting course to evade.

      This TED talk from 2 years ago shows that the Google car is capable of detecting and reacting to numerous different obstacles it has not seen before in an urban environment such as ducks and a elderly woman in a wheelchair chasing them on the road. The car halts, waits for them to pass and continues. And keep in mind this footage is old, the capabilities at this point have probably improved.

      Humans are on average really bad drivers because we cannot focus on very many things at once and our reaction time is relatively slow. Self-driving cars have both better awareness (ie. more sensory data) and the ability to continuously asses its surroundings in a way no human can, as well as having the ability to do math before making decisions. The car can calculate whether or not an emergency breaking maneuver will be enough to stop the car before it hits an obstacle, or if it's likely that the distance to said obstacle is too short and evasion is a better option, and it can calculate all of this in the second or so that it takes for your brain to even react to what's happening.

      I've driven in heavy-snow conditions here in Finland with very poor visibility and had some close calls. Would I trust a car such as Google's that has LIDAR capabilities and is thus able to sense objects in the distance before humans can even see them in such conditions moreso than myself? Yes. Why the fuck wouldn't I? There is absolutely nothing miraculous about humans which would make computers surpassing us in driving abilities impossible. In fact in a sense they've already done so as the accident rate for Google's car for example is far below human drivers, although obviously it will take time before the cars learn to handle different weather/road conditions, but there's no computational/technological reason why they won't be able to do so.

      --
      "It is the business of the future to be dangerous" -Alfred North Whitehead
    4. Re:I really don't understand the interest here by hipp5 · · Score: 1

      And what happens when a deer decides to bolt out from the woods in front of your vehicle? Are you going to trust that the car can detect a deer?

      I'm going to trust the car to detect the deer a lot more than I trust myself to detect a deer. I only have two eyes and I can only focus them on one point at a time. They also only see in the visible spectrum. My car could conceivably have 360 degree, multi-spectrum "vision".

    5. Re:I really don't understand the interest here by Kiuas · · Score: 1

      Replying to myself because I accidentally copied the wrong link which is to a clip of the said talk and not the whole of it. Here's the original TED talk. The part about the lady & ducks is slightly after 11:10.

      --
      "It is the business of the future to be dangerous" -Alfred North Whitehead
    6. Re:I really don't understand the interest here by be951 · · Score: 1

      And what happens when a deer decides to bolt out from the woods in front of your vehicle?

      Most likely, the computer system detects it more quickly than a person would, reacts faster than humanly possible, and brakes and/or steers in an optimized way to avoid both collision and loss of control.That kind of scenario is one where automated systems easily beat humans. The concerning ones are when visibility is poor, or lane markings are bad or confusing, such as in inclement weather or construction zones.

      just hire Disney's engineers and building a fucking monorail in most cities and connect them to the suburbs

      A train line is great if all the places you want to go are in a nice line. And we could certainly use better mass transit many places in the U.S. But it's only a partial replacement for driving, particularly with most cities/metro areas being laid out (or sprawled out, as it were) with cars in mind as the primary mode of transportation. Self driving vehicles are a more complete solution (even if they fall short of 100%) and much more cost effective when you consider that developing self-driving vehicles involves perhaps a dozen companies* conducting billion or multi-billion dollar projects vs. 5-10 dozen metro areas** (just in the U.S.) conducting multi-billion to tens of billion dollar projects to deploy light rail that would cover a fraction of the population over more limited use cases.

      For better or worse, the U.S. has a car- and truck-centric transportation culture. If we were starting from scratch, it might make more sense to do things differently. But as it is, we have to work from the infrastructure we already have. Which means making driving better. A good way to do that is to remove humans from the equation as much as possible since human error is the cause of nearly all traffic accidents.

      Plus, lots of people need to drive on a regular basis (daily commutes, etc...), and generally don't enjoy it. So having someone (something) else do the driving is an upgrade for most people. Especially if there is no trade-off with the other up aspects of driving -- departing when you choose, making additional stops as desired, privacy and personal space, ability to carry around an amount of stuff that would be prohibitive on public transit, and store things in the vehicle, etc....

      * With other companies buying components or licensing tech from the makers that do the best
      ** The U.S. has 382 Metropolitan Statistical Areas, over 100 of which have at least half a million population, and more than 50 of which have a million or more.

    7. Re:I really don't understand the interest here by Anonymous Coward · · Score: 0

      You do realize that all of those computations are like "add R1 to R2 and store in R2"? Computers can only handle that which they were programmed to handle. If the programmer didn't think of a scenario, then the computer won't know what to do. A computer will handle a situation identically every time. And if it doesn't know how to do something, it won't know every time very quickly. At least a panicky human can make something up. I don't know, maybe it's because I'm a software engineer, and have seen how the sausage is made, but this blind trust people have in computers is truly baffling to me. Seriously, pay real good attention to how often computers screw up. It's terribly frequently. I guess since I've been involved with the release candidate meetings where we're going through the bug list and deciding if any of these bugs are severe enough to block release that shakes my trust in the all mighty computer. The world is a complex thing, you're making thousands of decisions every minute even if you don't realize it. And a computer has to be programmed to handle every one of those. If ever there's one that it wasn't prepared for, nobody knows what will happen. The only thing we can be sure of is that the computer won't adapt.

    8. Re:I really don't understand the interest here by Anonymous Coward · · Score: 0

      p>Am I going to trust that a computerized system is capable of detecting an object in front of the car an apply brakes faster than a human being?

      Hell yes I am.?

      If detecting objects in the road were the only requirement then life would be a lot simpler for AD developers.

    9. Re:I really don't understand the interest here by swilver · · Score: 1

      The scenario is obstacle on the road, large enough to cause serious damage. It won't even bother to classify it as a deer, bear, wolf, whatever.

    10. Re:I really don't understand the interest here by fluffernutter · · Score: 1

      So it's going to stop for a large garbage bag blowing in the wind then? Wonderful.

      --
      Laws are rules for the court, but merely a bottom bar to hit for life. Think beyond laws in your actions always.
    11. Re:I really don't understand the interest here by aaarrrgggh · · Score: 1

      Depends on how you look at the efficiency of the solution: in terms of time all these Elon musings are more efficient; in terms of resource utilization, mass/rapid transit is more efficient. The caveat is you need walkable communities on either end for mass transit to work, and Elon isn't too close to a metro station.

    12. Re:I really don't understand the interest here by Rei · · Score: 1

      That's actually part of the problem with LIDAR - it's superb at seeing obstacles (in good weather conditions, at least), but it provides you with no information about what you're seeing. Everything is an obstacle.

      Radar on the other hand wouldn't even see the trash bag. But a bit of aluminum foil will give a huge blazing return.

      Camera vision systems might recognize a trash bag as a trash bag, but only if they were trained to and are proven good at their task.

      You really need a combination of sensors, and some very challenging software behind it that is trained to deal with a huge variety of edge cases.

      --
      The most exciting phrase to hear in science, the one that heralds new discoveries, is not âEureka!â(TM), but
    13. Re:I really don't understand the interest here by Immerman · · Score: 1

      Probably - as should you. Hitting that bag has a fair chance of getting it snagged on your car and obstructing your view. Probably don't even need to stop - just slow down a bit so it makes it across the road before you reach it, and so that you will have time to stop if you mis-identified it at first glance.

      --
      --- Most topics have many sides worth arguing, allow me to take one opposite you.
    14. Re:I really don't understand the interest here by Immerman · · Score: 1

      The problem is you're a software engineer - we tend to think like engineers - i.e.software is a bunch of deterministic sequences created by an engineer. Modern AI doesn't really work that way, which is why AI is a whole separate research branch. It probably has more in common with an actual neural circuit than traditional software, and it will generally *not* handle a situation identically every time - because situations are never identical to begin with.

      Now, perhaps they only use AI to categorize the environment, and then use that as the input to the driving algorithm, in which case you'd be correct. Even then though, it's a pretty limited problem space - in descending order of importance: "1) don't hit anything", "2) stay on the road", and "3) maintain forward progress at specified speed" for starters. And, if you're doing your job well "continually assess offroad situations in case you need to suddenly violate (2) to obey (1).

      Frankly, in a crisis situation the car is likely going to assess the obstacles, velocities, and spacings a lot more accurately than a human driver, as well as having a more accurate understanding of its handling characteristics under non-standard accelerations (such as skids and powerslides). And it's not subjected to the ~1/8 second signal propagation lag of the human body. If it has identified obstacles correctly enough, then it's almost certainly going to do a considerably better job of collision avoidance/minimization than you are. In fact it's a far easier problem than navigating city traffic, one almost entirely constrained by reaction times and physics - both areas where computers far outpace humans.

      Now, of course there's is always the chance the AI has egregiously misidentified an obstacle, or worse, failed to recognize one at all, and *that* could be a serious problem. In fact, I believe that's responsible for most autonomous-vehicle accidents to date in non-crisis scenarios. In a crisis scenario though, the many benefits will almost certainly save a lot more lives (and property damage) than sporadic recognition failures will cost.

      --
      --- Most topics have many sides worth arguing, allow me to take one opposite you.
    15. Re:I really don't understand the interest here by Anonymous Coward · · Score: 0

      Oh thank you so very much for pulling out the "AI" card. I knew somebody would say "it's fine because AI" when they clearly have no idea how AI works. Because as a software engineer who is experienced, I'm familiar with AI. What you just said is "We have software that is still going to act in a consistent manner, but now we don't know how to figure out what it's going to do or how it's gong to act". AI doesn't figure things out, it's got an internal model, it's going to throw inputs into that model, and we as humans just hope that it gets it right. If you want fun, look at the Google AI that they use to categorize images. Now, stop being impressed by how often it gets it right. That's unimportant. Pay attention to how often it gets it wrong. When you stop letting yourself get wowed by how often it's right, and pay attention to how often it gets it wrong, you start noticing it's quite often. I personally loved when it categorized a picture of a bear as a dog. These are two things you don't want to confuse. And it's not a rare occurrence. I remember recently somebody feeding a picture of a helicopter cockpit into it and it returned that it was a lake. This is the unpredictability of AI. And you want to trust your life to this? And this is literally one of the most advanced computer vision systems on earth today. And remember, I feed that picture of a helicopter at it, every time it'll see a lake until an engineer goes in and corrects it.

    16. Re:I really don't understand the interest here by Anonymous Coward · · Score: 0

      You've just proven the point.

      Based on how a garbage bag may be moving on the road and how the vehicle ahead may have affected its movement, a human will typically very quickly recognize whether it is OK to hit the bag or if it must be avoided. A human is more likely to accept the risk of hitting the garbage bag. And if hitting the bag does result in an obstructed view, any decent driver should be able to navigate their vehicle to the side of the road, should be able to judge by feel when they have reached the side of the road, and should be able to quickly and safely come to a stop.

      A computer simply cannot match a human in identifying that this rather large black mass that it is approaching at high speed is just a simple garbage bag. This would require true AI, without which the computer cannot possibly be allowed to make such a risk assessment and make the decision to hit the bag. The computer can only be programmed to avoid hitting it by either braking or navigating around it. At highway speeds, slamming on the brakes could be fatal for the occupants of the computer-driven vehicle as well as those in nearby vehicles. Veering around it is not any better - while the computer may be capable of making a very quick and accurate maneuver while simultaneously avoiding other vehicles in its vicinity, such a maneuver at highway speeds could cause nearby drivers to panic and attempt to avoid what probably appears to be an idiot driver suddenly jerking the wheel, likely resulting in an accident.

    17. Re:I really don't understand the interest here by Anonymous Coward · · Score: 0

      What you are describing is not true AI. Despite all the marketing bullshit spewed these days, there is still currently no such thing as true AI. You cannot possibly be familiar with AI because it simply does not exist at this point in time.

  5. exaspe by Anonymous Coward · · Score: 0

    don't even mention

  6. Bravo Toyota! by Anonymous Coward · · Score: 1

    Someone is actually taking time to think this through. Don't get me wrong. I'm ready for our self driving car overlords. I just want to make sure they are ready for the job first.

  7. Caution is important by GWBasic · · Score: 3, Insightful

    I'm a little skeptical of a sudden mass takeover with autonomous driving. As this post implies, the risk is huge. Where are autonomous devices in low-risk situations? Why haven't they taken over? I think we're better off with things like dryers that can sort and fold laundry, or dishwashers that can put the dishes away. The risk of a dropped dish or torn shirt is much more tolerable than a car crash at highway speeds.

    1. Re:Caution is important by hipp5 · · Score: 2

      The risk of a dropped dish or torn shirt is much more tolerable than a car crash at highway speeds.

      But the rewards are also so much lower.

      The potential rewards of autonomous driving are HUGE. I won't pay more than a couple hundred bucks to have my dryer fold laundry. But I would pay a lot to not have have society ever have to face drunk drivers again. To give old people the freedom to get out and about. To have cars that can precision park themselves, thereby taking up way less parking area (no door opening space needed). And so on.

    2. Re:Caution is important by atrex · · Score: 2

      Fabrics are actually very difficult for robotics to deal with because it bunches, snags, etc. It's why the manufacture of clothing is still mostly done in sweatshops instead of being completely automated.

      A dishwasher could be automated, but you'd be talking about having it integrated into a cabinet system and having grippers on slide tracks trying to grab non-metallic, non-magnetic plates and glasses with just the right amount of force not to break them and organize them throughout the cabinet. Any potential convenience would be grossly out weighed by the cost to develop or even manufacture and install the unit.

      There is a ton of automation already out there in "low risk" environments. The average consumer doesn't see it because the cost of it limits it to factories, shipping centers, and other large facilities.

      Arguably, the "high" risk nature of driving is one of the main reasons behind autonomous vehicle development. The developers believe that they can and will ultimately create self driving vehicles that are safer than the average human. And keep in mind that the average of human drivers includes those who are sleepy, sick, intoxicated, staring at their phone, road raging, etc.

      Of course, the other big factor in the research and development of autonomous vehicles is the shipping industry. How much can corporations save by replacing truck drivers with computers? Computers that don't need to sleep or eat and can be on the road traveling from point A to point B 24 hours a day 7 days a week. That's the jewel at the end of their rainbow.

      Unfortunately, there isn't going to be any sudden shift to fully autonomous vehicles, no matter how safe they may end up becoming. The change over is going to be long and arduous, simply because not everyone can just run out and buy a new car. Nor can most existing vehicles be easily retrofitted. I'd wager we're looking at a 30-50 year time span for the development of the vehicles and supporting infrastructure to finish and for them to trickle out through society.

      There's also a large potential for fully automated vehicles to cause disruption in individual car ownership. Cars may not be something everyone owns in 20-30 years. Instead, entire fleets of automated vehicles run by Uber, Lyft, or another startup could have their services provided to individuals via subscription fees and shuttle them here, there, and everywhere all with a few smartphone swipes.

    3. Re:Caution is important by Leuf · · Score: 1

      The thing is people are really quite bad at driving. People who are bad at folding laundry just have wrinkled clothes. People who are bad at driving kill people. "According to the World Health Organization, road traffic injuries caused an estimated 1.25 million deaths worldwide in the year 2010." How many technological advances have the potential to save a million lives per year?

    4. Re:Caution is important by GWBasic · · Score: 1

      There is absolutely no proof that a computer can drive a car better, and safer, than a human. None, whatsoever. Until the proof exists, I'd rather risk something that's cheap to replace instead of my life.

  8. Delusional by kackle · · Score: 1

    I hope your loved ones don't get harmed by this newfound laziness hiding behind flashy, imperfect technology.

    1. Re:Delusional by zifn4b · · Score: 1

      I hope your loved ones don't get harmed by this newfound laziness hiding behind flashy, imperfect technology.

      This is why I work from home. :) It's also better for the environment...

      --
      We'll make great pets
    2. Re:Delusional by Baron_Yam · · Score: 1

      >This is why I work from home.

      As someone who has had an (abysmally bad) driver determine that their home was a valid roadway... I question the value of your choice.

    3. Re:Delusional by Anonymous Coward · · Score: 0

      I hope your loved ones don't get hurt by a drunk or inattentive driver, or some other thing that happens too quickly for a human to react to where a computer might have a chance. As great as it would be for you to have an impetus to reflect on what a fucking retard you are, I don't want people to have to get hurt. Unless it's you. That would be fine.

    4. Re:Delusional by zifn4b · · Score: 1

      >This is why I work from home.

      As someone who has had an (abysmally bad) driver determine that their home was a valid roadway... I question the value of your choice.

      I'm not sure I follow your logic there, it seems to be nonsense. First of all, if you want to objectively make a claim the my choice of working from home is less valuable than commuting, then by all means lay out a claim and supporting logic and evidence for the claim. If your claim is subjective, meaning that you would not find my choice valuable, then we just agree to disagree because what is good for you is not necessarily good for me and vice versa but we should respect each other.

      --
      We'll make great pets
    5. Re:Delusional by Baron_Yam · · Score: 1

      >I'm not sure I follow your logic there, it seems to be nonsense

      I'm not sure I follow the motivation behind your post, but I'm guessing you have no sense of humour and a stick lodged up your backside.

  9. Re:A computer is needed to slam down the accelerat by mjwx · · Score: 1

    I remember way back when that there was a site that plugged itself as a "Toyota Simulator" (seems to be down now). When you went to the site, it was nothing but a looping video of a car driving way too fast down a road and a person screaming. ;)

    Compared to the Peugeot simulator, which is a car travelling down a stretch of road way too slow with the driver complaining about their prostate, hooligans, being half blind and the Yoofs of today.

    --
    Calling someone a "hater" only means you can not rationally rebut their argument.
  10. Loong hand-over times by DrTJ · · Score: 4, Insightful

    Toyota is not the only one deliberating skipping L3 and go directly to L4. Volvo intends to to the same, as well as some of the German vendors.

    The reason is that studies show that hand-overs do not only take "a few seconds" according to the article, but that there is a tail of up to 40 seconds before the "driver-to-be" comprehends the situation.

    Since 40 seconds is an eternity in traffic, it poses essentially the same challenges as L4 systems. So why bother with L3?

    1. Re:Loong hand-over times by Anonymous Coward · · Score: 1

      Bingo. Even highly-trained airline pilots have proven unable to deal with a handover from the autopilot to the pilots without flying a perfectly good airliner into the sea. So expecting the average driver to do it when they don't have minutes to react is crazy.

      What makes it doubly bad is that the computer will only hand over when it runs into a situation it can't handle, which means the human will only be expected to take over when the car is in a complex and dangerous situation to begin with.

    2. Re:Loong hand-over times by swillden · · Score: 1

      Toyota is not the only one deliberating skipping L3 and go directly to L4. Volvo intends to to the same, as well as some of the German vendors.

      Google (now, Waymo) also decided years ago that L3 is a bad idea.

      --
      Note to ACs: I usually delete AC replies without reading them. If you want to talk to me, log in.
    3. Re:Loong hand-over times by sl3xd · · Score: 1

      Even with the semi-autonomous systems in my car (adaptive cruise control and lane guidance), hand-off is a big problem. Either the car decides to slam on the brakes when a car starts to pull off the road (slows down and switches lanes), or it accelerates with reckless abandon because it can't sense the stopped cars 300+ feet ahead. (And it won't slow down when it does detect the cars ahead because the speed delta is now greater than 30 MPH)

      Or it starts tugging at the wheel to guide my car back into the center of my lane, even though the reason I'm departing the lane is because I'm being crowded out of my lane.

      It feels somewhat like teaching somebody to drive... the system almost always does the right thing, but sometimes makes decisions you don't agree with, and the transition between who is in control is jarring.

      --
      -- Sometimes you have to turn the lights off in order to see.
  11. If they won't be first by Anonymous Coward · · Score: 0

    Let's look at Volvo.

  12. It IS unsafe by Anonymous Coward · · Score: 1

    Expecting the human to take over in a panic situation IS unsafe. The human should only be taking over while parked.

    There is exactly one class of people who have the training qualifying them to take over a driving car: Driving instructors. And they are usually limited to stepping on the brake, something the autonomous car could easily do on its own in a panic situation.

    However, that's not saying that we have to go straight from level zero to level five. We just have to do it in a different way.

    Rather than letting the car drive on the straight road, and expecting the human to take over in case the car overlooks a pedestrian, we should be letting the human drive on the straight road, and let the car take over when the human overlooks a pedestrian.

    Because computers are really good at two things: Monitoring without getting bored, and making split-second decisions. Two things that humans are quite bad at.

    If we started there, we could expand the number of situations where the computer can take over until it can do the entire driving by itself. The current plan is to reduce the number of situations where the human must to exactly what humans are bad at until the car can handle everything on its own.

    1. Re:It IS unsafe by Wycliffe · · Score: 1

      Expecting the human to take over in a panic situation IS unsafe. The human should only be taking over while parked.

      I agree completely. Doing a handoff of a moving car is just asking for trouble.

      Rather than letting the car drive on the straight road, and expecting the human to take over in case the car overlooks a pedestrian, we should be letting the human drive on the straight road, and let the car take over when the human overlooks a pedestrian.

      There is a 3rd solution which I prefer. Let the computer drive on straight roads in good weather on limited access highways first. i.e. the boring stuff. If the weather starts to turn bad or you are approaching a city, have the computer pull over to the side of the road so that you can switch driver. This is already common practice. Growing up on vacations, my mom would help drive on the long stretches and then pull over and let my dad drive when we got to a big city or close to our destination.

    2. Re:It IS unsafe by Anonymous Coward · · Score: 0

      There is exactly one class of people who have the training qualifying them to take over a driving car: Driving instructors.

      What exactly would be the problem with introducing an additional license requirement for operating vehicles on auto-pilot?

      (inconvenience of the US style DMV and low'ish requirements to get a license in the first place aside, of course)

    3. Re:It IS unsafe by Anonymous Coward · · Score: 0

      Most people want the burden of driving decreased, if you need to be more qualified to operated a self-driving car, it'll probably be more effort than just driving it yourself. Which defeats the point of them.

  13. underground tunnels cost way more then truckers by Joe_Dragon · · Score: 1

    underground tunnels cost way more then truckers

  14. Search and Replace by Anonymous Coward · · Score: 0

    It was nice of the Toyota administrative assistant to search and replace all instances of "liability concerns" with "safety concerns".

  15. Why isn't there an African car company? by Anonymous Coward · · Score: 0

    Anybody? Any answers? I just can't imagine...

    1. Re:Why isn't there an African car company? by Anonymous Coward · · Score: 0

      There are.

      https://www.brandsouthafrica.com/investments-immigration/africa-gateway/8-african-cars-270116

  16. I'm sick of hearing this BS by Viol8 · · Score: 2

    "Humans are on average really bad drivers"

    No, we're not. We're actually bloody good at it considering we never evolved to drive something weighing 1.5 tons at anything up to 10 times our maximum running speed alongside other vehicles doing similar speeds.

    People cite accident statistics as if they're significant. When you consider the TRILLIONS of miles driven every year by the worlds drivers and the number of potential accidents that DIDN'T happen because drivers reacted properly, the actual number of accidents is a statistical blip.

    If you think computers are so much better at these sorts of tasks than humans then you might like to consider why Airbus computers will hand over control to the pilot if they're unable to handle the situation. And these are computers in a $200,000,000 aircraft, not a $20,000 car!

    1. Re:I'm sick of hearing this BS by fluffernutter · · Score: 1

      ..which are also operating in a controlled airspace with much less happening from second to second than on the ground. Also monitored by air traffic control.

      --
      Laws are rules for the court, but merely a bottom bar to hit for life. Think beyond laws in your actions always.
    2. Re:I'm sick of hearing this BS by apoc.famine · · Score: 1

      And the pilots have trained for that exact scenario, and have drilled on it, and have studied the common causes of that. Yeah, the comparisons to a car are not really useful here.

      --
      Velociraptor = Distiraptor / Timeraptor
    3. Re:I'm sick of hearing this BS by Viol8 · · Score: 1

      I think you rather missed the point - the computers in aircraft have an easier time than would computers in a completely automated car, yet they still have to hand back control to the pilots from time to time.

  17. Good, rather not see Level 3 at all... by EndlessNameless · · Score: 1

    Whenever you have to hand control of the vehicle back to the human, there is going to be a delay. This is absolutely unavoidable and potentially very dangerous.

    The driver, who was presumably inattentive during the fully-automated drive, will have to assess the surroundings and respond. This makes existence of a SAE Level 3 car inherently unsafe---there is little empirical support for idea that we can have a safe sometimes-automated system that fails into manual control.

    Human attention change, perception times, and decision-making times are all fairly well understood now. Individuals fall into a reasonably large range, but even the best-performing humans will have trouble asserting control in a timely fashion---especially if the automated system is standing down due to an ambiguous and potentially dangerous condition. Snap judgements and muscle memory are barely enough

    While I applaud the progress, I believe that Level 3 autonomous cars should never be sold as such to the public. The manufacturer can test these systems and gather all the data they want, but the driver should be required to maintain situational awareness at all times.

    --

    ---
    According to the latest ruleset, this post should be modded as Vorpal Flamebait +5.
  18. Re: underground tunnels cost way more then trucker by zifn4b · · Score: 1

    underground tunnels cost way more then truckers

    I think you would have to qualify this a little better. Sure, the up front cost is substantial. However, being that a logistics company would not have to pay for drivers at CDL hourly rates anymore, eventually it would pay for itself and then pay dividends.

    --
    We'll make great pets
  19. Brakes? by Marqis · · Score: 1

    Considering Toyota couldn't write software for brakes I'm not surprised they're uneasy about writing autonomous driving software.

  20. Experience is critical to safe driving by Sqreater · · Score: 1

    So, after a while with cars driving themselves, what exactly will the turnover to the now rusty human do but insure a crash? Also, there is not instantaneous human situational awareness. What...wait...oh, THAT is happening, so I must.....crash. Driving is a never ending story. You must pay attention until you turn the key to off. Also, the system is perfect, or it is not. Everything degrades and eventually needs repair. And how dangerous is a degrading self-driving-car system? And, since they cannot legally program their cars to do the complex conspiratorial law-breaking we all do to maintain the traffic flow, life itself will slow down. Commerce will slow down. That will not be acceptable to individuals or businesses. Thus, either cars will be allowed to break the traffic laws, or the traffic laws will be rewritten to accommodate them. But if they break the traffic laws as they exist willfully, car companies are open to massive insurance hits. And even criminal conspiracy charges are possible. If they massively rewrite the traffic laws of the nation, how long will that take? And how, exactly would they do that across all the States? The whole autonomous cars thing is a deadly dog's breakfast waiting to be served up to a helpless public by a greedy industrial base and a corrupt and incompetent government.

    --
    E Proelio Veritas.
  21. This will end in disaster by Rick+Schumann · · Score: 1

    Until we have REAL AI, self-aware, capable of actual thought and real interaction with humans, and not the current dead-end approach ('learning algorithms', 'expert systems', etc) we will not have truly safe, effective self-driving cars -- and having these half-assed systems fully in control of a vehicle, with no possibility of humans taking control, we will have death and disaster. At best we'll have totally frustrating vehicles that stop for apparently no reason, while it 'phones home' so a human operator can remotely control it, which is also totally unacceptable. You can keep your so-called 'self driving cars', because I don't want to die. Furthermore we won't have REAL AI for a long time to come yet, if ever, because we have no idea how an actual brain is conscious, self-aware, and capable of true ability to think. It's all marketing hype and nonsense perpetuated by people who have no clue what's really going on, and they can keep it, I'd sooner WALK everywhere.

    1. Re:This will end in disaster by sl3xd · · Score: 1

      Until we have REAL AI, self-aware, capable of actual thought and real interaction with humans

      We don't even know what "self-aware" is, to say nothing of "actual thought". How exactly is it defined? A sea cucumber is aware enough of itself to try to preserve its existence, and yet they have no brain. Is that "self aware?" They're not bumping around in the dark, speeding off in random directions, randomly eating, or mating with whatever they touch. They can process input from their sensory organs to find food, and are able to communicate for the sake or reproduction. Is that thought?

      How is a sea cucumber's intelligence any less real than our own basic "can it kill me, can I eat it, can I hump it" model?

      Moreover, the standard isn't that an autonomous vehicle is fully self aware and of a human level of intelligence, and in fact, that kind of awareness may be detrimental to autonomous driving - an AI of even dog levels of intelligence may be easily distracted by squirrels.

      Autonomous cars merely have to be better than human drivers, which is much simpler, because humans are honestly bad drivers - we make profoundly unsafe decisions based on any of a number of things that have nothing to do with driving the car - such as thinking about anything except driving the car. We convince ourselves we're paying attention, yet we've refined the ability to prove the audience isn't paying attention into an artform we pay to see (just ask Penn & Teller...)

      So no, autonomous cars don't have to be a superintelligent AI in order to be a better choice than our current insane, hormonal ape model.

      --
      -- Sometimes you have to turn the lights off in order to see.
    2. Re:This will end in disaster by Rick+Schumann · · Score: 1

      We don't even know what "self-aware" is, to say nothing of "actual thought". How exactly is it defined?

      ..and THAT is why we can't create REAL AI; we have NO IDEA how things actually work, and you are taking for granted how complex a task driving is, which is why you need a mind that can actually THINK.

    3. Re: This will end in disaster by Anonymous Coward · · Score: 0

      Nowhere near as much âoethoughtâ is required as you seem to believe.

      With new (in development) GPS receivers already able to get centimeter-accurate positioning, the task is becoming even simpler.

      Stereoscopic cameras and LIDAR are only two of several ways that are adept at generating point clouds â" which are then used for positioning and collision avoidance.

      Similar point cloud data is already whatâ(TM)s used in modern driving games - including full route pathing and reacting to the crazy ape playing the game. There are even competitions to build better driving AIâ(TM)s (TORCS being a much-used example).

      I worked with autonomous robots nearly 20 years ago that could navigate busy parking lots without problems, and it used less computing power than a smart watch.

      In short, expecting perfection is unreasonable, as the only (non-disease) thing that kills more Americans than our incompetence at driving is suicide.

      Think about that for a second.

  22. I'm just happy to have blindspot monitoring by Anonymous Coward · · Score: 0

    I basically was terrible at shoulder checking for years and years and now I never really have to. The rest of the tech toys I can do without. The lane keep assist stuff is just annoying and the auto breaking thing is a great way to get rear ended. I disable both when driving. But man do I love blindspot monitoring systems. Now I can change lanes with confidence knowing that the worst that can happen is I crush some motorcyclist with my SUV.

  23. Do it like the Air Force by Anonymous Coward · · Score: 0

    Air Force has word phrasing to switch between pilots.

    "I have" "You Have"

    So a strong alert tone, and "You Have" in a booming James Earl Jones type voice should be enough...

  24. These cars dont need to be as good as humans by Anonymous Coward · · Score: 0

    Self driving cars don't need to be as good at driving as humans to be vastly safer than humans and to minimize traffic. For example just by eliminating the option of passing on the right and minimizing or eliminating the use of left turns they will be vastly safer than human drivers who routinely pass on right when they shouldn't and make poorly judged left turns when we have thousands of traffic studies that not only show how comparatively risky left turns are but also that they create traffic and actually slow down you time to destination (which is why UPS now orders their drivers never to make left turns). Google's self driving car doesn't turn left and never passes on the right. Those two things alone make it better than human.