back to article Google robo-car in rear-end smash – but cack-handed human blamed

Passengers of a Google self-driving car were hurt when the robo-vehicle collided with another car – but Google insists the human at the wheel of the other car is to blame. The web goliath has often blamed the human inability to manoeuvre vehicles when its own driverless machines have been involved in prangs. A blog post by …

  1. Paul Crawford Silver badge

    While this one example is clearly a not-paying-attention driver, Are there any stats to say if these events in total are more or less common than a meatsack driven car?

    1. NumptyScrub

      I'd suggest that these events are probably representative of driver experience in that locality. For us meatsack drivers, this is (literally) just another vehicle on the road, and will be treated as such by other drivers. Possibly a minority might recognise it as an autonomous vehicle, but anyone paying that much attention to be able to do so is (in my experience) the exception rather than the rule :'(

    2. Rafael 1

      "Harold, dear, you're getting too close to that weird looking car!"

      "Don't worry, Martha. It is one of those whatchamacallit Google Drive Carts, it will see my car and automatically get out of the way. Just watch!"

    3. JeffyPoooh
      Pint

      Ironically, the driver behind was...

      Ironically, the driver behind was checking her Gmail on her Android mobile.

    4. jviddy

      I'm guessing your question depends on whether these driverless cars over react in certain situations or respond that much more quickly than a human driver that the driver behind has significantly less time to react.

      1. Anonymous Coward
        Anonymous Coward

        > I'm guessing your question depends on whether these driverless cars over react in certain situations or respond that much more quickly than a human driver that the driver behind has significantly less time to react.

        Well if what we are lead to believe, at least in this particular case, the cars were stopped at the junction waiting to pull of. The car involved in the accident just didn't notice the stopped traffic and ploughed into the Google car.

        It likely happens all the time but there is greater publicity when a Google car is involved.

      2. John Brown (no body) Silver badge

        "I'm guessing your question depends on whether these driverless cars over react in certain situations or respond that much more quickly than a human driver that the driver behind has significantly less time to react."

        People are used to the reactions and foibles of other human drivers, not so much robots. Judging by the wire frame video of the situation, the Google car stopped quite some distance behind the car in front. Much more so than a normal human driver would do which may have contributed in that the following driver expected to stop further on and for whatever reason looked away at a dangerous moment. I'm not defending the driver who rear-ended the Google car, but the Google car seems to have stopped much sooner than most people would expect, especially during what looks like a busy period when cars bunching up at a junction mean more get through the green light and take less space up on the road.

  2. Trollslayer
    Terminator

    I welcome our automative overlords.

  3. Dazed and Confused

    "and they never get tired, irritable or distracted".

    Let's just hope they never BSOD

    1. Paul Crawford Silver badge

      Re: "and they never get tired, irritable or distracted".

      Lets hope the quality of code is better than the Toyota engine management unit:

      http://betterembsw.blogspot.co.uk/2014/09/a-case-study-of-toyota-unintended.html

      Read it an weep, or at least, not buy a Toyota.

      However, who knows how well other suppliers would fair if properly audited. Eh, Ford with your engine stopping problem, or you Range Rover with your door unlocking?

      1. ma1010
        FAIL

        Re: "and they never get tired, irritable or distracted".

        Thanks for this. You've convinced me to keep driving my '87 Toyota. The throttle in that vehicle is controlled ONLY by my foot on the pedal. Brakes, too. Just the way I like it. Guess I'll drive old cars until I stop driving.

        1. Destroy All Monsters Silver badge
          Facepalm

          Re: "and they never get tired, irritable or distracted".

          Toyota engine management ... 256.6K Non-Comment Lines C Source on page 18.

          Even if this is something lawyerly-arguably close to MISRA C .... WHAT!

          The inertia inherent in industry, bullshit whirling around in closed loops in execs' heads and the cruft pushed by "learning institutions" and impressed on young minds must be just insurmountable.

          1. Destroy All Monsters Silver badge

            Re: "and they never get tired, irritable or distracted".

            Safety rules typically forbid recursion: Risk of stack overflow, Recursion makes it impossible to do the V&V necessary for a system like the ETCS

            Even V&V consists of badly processed knowledge, applied badly.

            1. Vic

              Re: "and they never get tired, irritable or distracted".

              Even V&V consists of badly processed knowledge, applied badly.

              That depends on how you do it.

              I've worked in environments that require proper V&V. I've yet to have a plane drop out of the sky because my code screwed up. But it is expensive...

              Vic.

          2. Vic

            Re: "and they never get tired, irritable or distracted".

            Even if this is something lawyerly-arguably close to MISRA C .... WHAT!

            No - it simply means there are 256K lines of C left once you've stripped the comments out.

            This is a reasonable way of determining the amount of code involved - the number of comment lines varies markedly between different programmers.

            Vic.

    2. hplasm
      Holmes

      Re: "and they never get tired, irritable or distracted".

      Let's hope they never run Windows, then.

    3. VinceH
      Terminator

      Re: "and they never get tired, irritable or distracted".

      obTerminator:

      "And it absolutely will not stop, ever*, until you are dead at your destination."

      * Except where required to do so in accordance with local traffic regulations. Or in the event someone crashes into it.

  4. fandom

    "People, on the other hand, appear to be bloody ignorant apes."

    Was there ever any doubt?

  5. Cari

    Man, I hope Google checked their cars break lights are working.

    1. I ain't Spartacus Gold badge
      Happy

      Is that the light that lights up when you have a KitKat?

    2. Anonymous Coward
      Anonymous Coward

      "Google checked their cars break lights are working"

      I hope they have a light that comes on to say that it's broken !

    3. This post has been deleted by its author

    4. Anonymous Coward
      Anonymous Coward

      > Man, I hope Google checked their cars break lights are working.

      The Google car was stationary, so chances are the brake lights would not have been lit. It is only people, who don't know how to drive that keep their foot on the brake when stopped.

      1. Will Godfrey Silver badge
        Thumb Down

        On the contrary. You are advised to keep your foot on the brake until the car behind has stopped. Especially if you are the front car at an intersection!

        1. heyrick Silver badge

          Here in France

          It is common practise in this situation to put the hazard warning lights on if you are the last car stopped. If somebody comes up behind you, you turn yours off and it becomes their problem...

      2. Terry Barnes

        Or, erm, the vast majority of US drivers in automatic cars. US driving practice is different to British practice.

      3. Joe Harrison

        Totally agree. Brake lights are (intentionally) very bright and I really don't need them shining in my face for a couple of minutes because the guy in front could not bother to engage the handbrake like the highway code says you should.

        1. Will Godfrey Silver badge

          Not been on one of the driver awareness courses then have you?

          Yes, I was speeding last year. 35MPH in an area that had recently switched from 40 to 30.

        2. Eddy Ito

          ... because the guy in front could not bother to engage the handbrake like the highway code says you should.

          Like the highway code where? We have at least 51 different highway codes to go with 51 different driver's licenses. Granted none are very different but having had a license in several states over the years, none have really mentioned the [hand, parking, emergency] brake other than to say it needs to be in working order. D.C. requires that it be accessible by the examiner when you take the driving test but that's not always the case.

        3. Orv

          I've never heard of this in the U.S. In fact I was encouraged in driver's ed to keep my foot on the brake while stopped, so traffic coming up behind me would see the brake lights and realize at a glance I wasn't moving. The only handbrake use covered was for parking. (I later learned how to use it for hill starts in a stick shift car, but driving stick is not taught in driver's ed in the U.S. because most cars are automatic.)

          Another benefit of keeping your foot on the brake is if you get smacked from behind, your car is less likely to shoot out into traffic.

          In parts of the U.S. that get snow and use road salt the handbrake usually stops working after a few years anyway, due to the cables seizing up. Sometimes they set and then fail to release, which leads one to become wary of them. My dad once bought a Ford that had it already disabled at the factory. ;)

          1. Vic

            Another benefit of keeping your foot on the brake is if you get smacked from behind, your car is less likely to shoot out into traffic.

            Unfortunately, it also tends to create a hot-spot on the brake disk. This can cause warping - particularly if you've just done a lot of braking (e.g. come of the motorway).

            Vic.

    5. Anonymous Coward
      Anonymous Coward

      "break lights"

      Sigh...

  6. Doctor Syntax Silver badge

    What do the statistics tell us?

    If the accidents are happening to Googlemoblies significantly more often than to conventionally driven cars then it does raise the question of what might be the explanation. What is different about the behaviour of these vehicles which contradicts the following drivers' expectations?

    1. hplasm
      Happy

      Re: What do the statistics tell us?

      "What is different about the behaviour of these vehicles which contradicts the following drivers' expectations?"

      They unexpectedly follow the rules of the road?

    2. sandman

      Re: What do the statistics tell us?

      "What is different about the behaviour of these vehicles which contradicts the following drivers' expectations?"

      They actually obey the rules of the road? ;-)

    3. Tenacal

      Re: What do the statistics tell us?

      While not an invalid question, I think it's a bit too soon to be asking. 14 'droid-meatsack collisions in 6 years against the large number * of regular meatsack-meatsack collisions isn't going to be enough to draw an analysis, even when weighted against the number of 'droid cars vs the amount of regular cars.

      *I have zero stats to back up this claim.

    4. John Robson Silver badge

      Re: What do the statistics tell us?

      Not sure they are - it was a few years ago that the google mobiles had covered more than the average driver does in their lifetime (well, assuming they drive from 17-87)

      There has been much more driving since then, so quite possibly two lifetimes so far? (pure guess) and they've been rear ended 11 times (and three other collisions). In this case it wasn't even the GCar that was the first to stop - the GCar had two vehicles ahead of it which had stopped!

      1. Anonymous Coward
        Anonymous Coward

        Re: What do the statistics tell us?

        "had covered more than the average driver does in their lifetime "

        But not the same sort of routes yet? Majority of British drivers could drive round the M25 without incident save some annoyance at the junctions near Heathrow, run the length of the A1, or park on the High Streets of the UK. Maybe survive the Champs Elysee. Have the google cars done anything equally challenging ?

        1. John Robson Silver badge

          Re: What do the statistics tell us?

          @AC - Have the google cars done anything equally challenging ?

          Can't quite work out if you think the M25 is challenging, or if you haven't seen any of the Google Car videos. They are tackling the fairly hard problem of city street driving, although they can do freeways those are pretty easy targets, very limited vehicle types and decisions to be made.

          In *this* case the car stopped (as would anyone else) because the vehicle in front had stopped... OK I rescind my "anyone else", because the driver behind didn't stop, or even slow...

          These cars are being tested in simulators, and presumably on test tracks, but they are also being tested in rush hour on a busy junction (sufficiently busy that the far side of the junction is blocked, so the correct response is to wait (as demonstrated by the human two cars ahead).

          1. Anonymous Coward
            Anonymous Coward

            Re: What do the statistics tell us?

            Two things about the M25 that are challenging to the less-experienced driver.

            Trying to get into or out of the nose to bumper crawl when it slows down past Heathrow. ie easing your nose forward until a driver decides he really has to let you through

            Changing lanes when the carriageways are full doing 55 mph, 70 mph and more like 85 mph respectively and few people are signalling.

            1. Anonymous Coward
              Anonymous Coward

              Re: What do the statistics tell us?

              I drove the M25 (and 4 other motorways) twice almost every weekday for 7 years. It is not a challenging motorway for driving on. It may sometimes be a challenge for a human having to crawl along it (although that was much rarer than suggested) but an automaton would love it and I could climb in the back and get some shut eye while it did.

              1. Archaon

                Re: What do the statistics tell us?

                "It may sometimes be a challenge for a human having to crawl along it (although that was much rarer than suggested)"

                That rather depends which bit you're driving on.

                From Chertsey (J11) to the M40 (J16) - and vice versa - is a crawl. A moving crawl, with a few short, sweet breaks in between queues, but a crawl nonetheless. During a weekday rush hour that 20 mile segment typically takes between an hour and 90 minutes - longer than the other 70-80 miles of my journey. That is pure congestion, and can be worse if there's an accident.

                That said, having mostly experienced the M25 westbound I was pleasantly surprised with the M25 eastbound yesterday (using the M1 at J21 as a reference point for east & west) - no problems whatsoever. Plenty of traffic still but free-flowing, able to do the speed limit etc.

                1. Alan Brown Silver badge

                  Re: What do the statistics tell us?

                  "During a weekday rush hour"

                  There's your mistake right there. If you're regularly encountering the same problem at the same time of day in the same area, experience would suggest altering your timetable.

                  1. Charles 9 Silver badge

                    Re: What do the statistics tell us?

                    "There's your mistake right there. If you're regularly encountering the same problem at the same time of day in the same area, experience would suggest altering your timetable."

                    And YOUR mistake is assuming people CAN alter their timetables. Many of these people are nine-to-fivers on fixed schedules (courtesy of their bosses) who need to be a point B within a certain amount of time or they'll miss something important that ruins the rest of their schedule.

                    So let's posit. If you're regularly encountering the same problem at the same time of time at all possible locations (which tends to happen in rush hour--people quickly learn the detours, others follow, and these get jammed, too) and you have no freedom to change your timetable, you only have one option: grin and bear it because you are just plain STUCK.

        2. DaLo

          Re: What do the statistics tell us?

          More challenging - which I have thought about is a section of road near me where in the morning all cars have to do a manoeuvre which is possibly illegal, if not it definitely doesn't follow the rules of the road.

          If you didn't do it though and were to strictly follow the highway code you end up backing up a whole lane of traffic and would be stuck there for a while.

          So would a robo-car be allowed to do an illegal manoeuvre (at certain times of the day), would this need to be a specific override programmed in or would it be able to use an artificial intelligence to say "everyone else is doing it I will as well - which could be very dangerous).

        3. Anonymous Coward
          Anonymous Coward

          Re: What do the statistics tell us?

          Having had daily commutes that include both the A1/M25 and the 101 in Mountain View (although not both at the same time, that would be a long commute) I'd say their current environment is harder.

          If there were no other cars on the road or they were all computerized the US roads would be far easier than the UK. The lanes are wider and straighter and you don't have to worry nearly as much about cars parked on residential streets effectively turning the road into a single lane for both directions. From a purely logical point of view lights are easier than roundabouts for a car to cope with.

          However in the US the traffic will unexpectedly stop far more frequently, people run red lights more (not counting the completely legal right turn on a red) and people in their size of a small house SUVs are notorious for not checking or signalling before changing lanes. I lost track of the number of times I had to swerve or slam on the brakes because someone tried to drive into the side of me.

          And at the risk of applying a massive stereotype different ethnic groups tend to have different driving styles that are not always compatible. In the bay area you have a population which has jumped massively in the last 20-30 years. There are lots of people from all over the US, the far east, India, Mexico and Europe, most of them moved there for work after they had learnt to drive in their local style. It makes the traffic a little less predictable than normal.

        4. Charles 9 Silver badge

          Re: What do the statistics tell us?

          "But not the same sort of routes yet? Majority of British drivers could drive round the M25 without incident save some annoyance at the junctions near Heathrow, run the length of the A1, or park on the High Streets of the UK. Maybe survive the Champs Elysee. Have the google cars done anything equally challenging ?"

          I would think negotiating a major Californian city like San Francisco or Los Angeles at rush hour would provide plenty of challenge. The cars have already crossed the Golden Gate Bridge, which has reversible lanes.

    5. JP19

      Re: What do the statistics tell us?

      "What is different about the behaviour of these vehicles"

      They probably dawdle but the biggest difference is they are small and look like something Noddy should be driving.

      In a momentary brain fart they look less like something you shouldn't drive into.

      1. John Brown (no body) Silver badge
        FAIL

        Re: What do the statistics tell us?

        "They probably dawdle but the biggest difference is they are small and look like something Noddy should be driving."

        Maybe El Reg is at fault for providing an incorrect "hero" image, or maybe you are at fault for commenting when you obviously don't have all the facts. It was a stonking big Lexus SUV, not a "Noddy car"

        "One of our Lexus vehicles..."

    6. Six_Degrees

      Re: What do the statistics tell us?

      Last time this issue came up (a few weeks ago) Google cars were having accidents per million miles at about half the average rate.

    7. Alan Brown Silver badge

      Re: What do the statistics tell us?

      "If the accidents are happening to Googlemoblies significantly more often than to conventionally driven cars"

      Except that:

      1: They're not

      2: The US highways administration estimates that more than half of minor collisions are never reported and freely admits this makes generating minor crash statistics virtually impossible.

      If this was a non-autonomous wagon whack the odds are good that it wouldn't have been reported, whereas Goo and Delphi are reporting everything from paint scrapes to near misses.

  7. Anonymous Coward
    Anonymous Coward

    Room for improvement

    Sounds the like the cars need to sense idiots on the road and do more to protect passengers once it knows a collision is inevitable - in this case do something to mitigate whiplash.

    1. DJV Silver badge
      Happy

      Re: Room for improvement

      "the cars need to sense idiots on the road"

      Add that in and they would all be nervous wrecks within a few hours!*

      Also, how about adding front and rear mounted indelible paint/ink guns to mark the crashing perp should they decide to try to make a quick getaway after the "attempt to mate" event?

      *especially if the software was written by the Sirius Cybernetics Corporation.

      1. Stoneshop Silver badge
        Go

        Re: Room for improvement

        Not just sense idiots, they also need to predict their unpredictable behavior. In other words, Defocused Temporal Perception.

        Just make sure there's room in the basement, next to the elevators.

        1. Alan Brown Silver badge

          Re: Room for improvement

          "They also need to predict their unpredictable behavior. "

          In my experience, most crashes are predictable. If the driver in front or behind is driving erratically, they're going to crash. It's just a matter of when and how messily.

          As mentioned, the Googlewagon could have predicted the impact and prepared for it, in the same way that many high end cars predict oncoming crashes.

          Anecdote: Stopped on the A29 behind about 15 other cars, I watched a driver coming up "rather rapidly" behind me, despite my brake lights going on. His nose dipped heavily as soon as the hazards went on and he just managed to stop without shortening my vehicle. Sometimes it's just a matter of making the guy behind break out of his white line trance.

    2. Six_Degrees

      Re: Room for improvement

      Like what? Pretty much all the driver can do to lessen whiplash is realize that a collision is imminent and take the brake off. That, though, has the downside of tossing the car forward, quite possibly into cross or oncoming traffic.

      There's a reason cars have those "headrests" - they're meant explicitly to reduce whiplash injuries.

      1. Orv

        Re: Room for improvement

        Wouldn't taking the brake off make it worse? Seems to me whiplash would be proportionate to how much the car accelerates, and a car with the brake on is going to accelerate less than one rolling free. The car might take more damage, though.

  8. Oh Matron!

    Crumple zones

    People are stupid, we all know this. People are also normally task focused, resulting in slight distractions actually taking attention away from the task at hand. Thus, People will always crash.

    Now that we've gotten that out of the way, as people will always crash, where the hell are the crumple zones on these things? I'm pretty sure that if your average america pickup truck would obliterate the car and entangled occupants easily, given how cheaply made these things look (no different from your average android phone, funnily enough)

    1. Archaon
      Facepalm

      Re: Crumple zones

      Sure, take a cheap shot at Android because Google are so well known for making the hardware for Android phones.

      Oh, wait...

  9. John F

    But did the google car take action to lessen the impact?

    In my one rear-ended collision like this I was able to see it coming - and 17mph is obvious - and remove my foot from the brake to lessen the effect on me. The google car could also have hooted when it saw the too-fast approach or moved. Did it? Not specified. That could skew the statistics, surely.

    1. Charles 9 Silver badge

      Re: But did the google car take action to lessen the impact?

      Releasing the brake is not an option if you're already in a queue (IIRC this was the case here; the Google car was third), as trying to take the impact this way simply careens you into the car in front of you, making two accidents in place of one.

  10. Little Mouse Silver badge

    "hit surprisingly often by other drivers..."

    Sounds suspiciously like the (very) old mother-in-law joke: She's never had an accident, but seen plenty of them in her rear-view mirror...

    1. nijam Silver badge

      Re: "hit surprisingly often by other drivers..."

      Well, the most dangerous component of a car is the nut that holds the steering wheel.

      As we're re-visiting old jokes that are actually amusingly-phrased statements of simple fact.

  11. knarf

    I ride motorcycles so...

    I treat all car drivers as complete idiots and this seems to be proved rather life saving on a number of occasions.

    I have noticed that if a car slightly wobbles then there is a good chance it will change direction, its if the driver has a little practice before doing it. Not sure silicon drivers will do me the same favour.

    1. fandom

      Re: I ride motorcycles so...

      I drive a car, so I treat all drivers, cars, motorcycles and bicycles, plus pedestrians as complete idiots.

      Not that I see myself as an exception.

      1. Alan Brown Silver badge

        Re: I ride motorcycles so...

        No matter how good a driver or rider you are, you will always make stupid mistakes or get distracted.

        Even the best drivers can't cope with more than 4-5 different obstacles and most have trouble coping with 2-3 (Eg, watching that wobbling cyclist and that guy swinging his nose wide whilst parallelparking on the wrong side of the road, completely missing the pedestrian crossing just past the parking car and the pregnant lady who's just stepped out onto it without looking.)

    2. Bert 1

      Re: I ride motorcycles so...

      You can add to to the list:

      Driving at the edge of the lane when in traffic queues.

      Driving too slowly

      Driving too quickly (and I don't mean just over the speed limit)

    3. Anonymous Coward
      Anonymous Coward

      Re: I ride motorcycles so...

      "Not sure silicon drivers will do me the same favour."

      They will, however, use their indicators properly, which is better.

      As a bit of a robotic driver myself I used to enjoy driving in Switzerland and North Germany where people might drive fast but they obeyed the rules. After the UK it was amusing to watch cars emerging from a tunnel on the other side of the road and their headlights being switched off immediately as if there was some transmitter in the road automating the process.

  12. Anonymous Coward
    Anonymous Coward

    "and they never get tired, irritable or distracted".

    I had visions of Sarah Connor looking at a terminator keeping watch, while taking refuge in a garage when I read that.

  13. Elmer Phud

    Fun and games

    I can't help but wonder what japes a group of cyclists can get up to by surrounding one of these things and riding alongside.

    If we can ride around town developing techniques for dealing with meat-controlled vehicles, how might these be use when faced with auto-boxes?

  14. 2Fat2Bald

    not being funny. But if 3 people go to hospital in a rear-end-at-17mph collision, then the car isn't very crashworthy. I'm assuming a genuine need and not just a precaution, of course.

    I wonder. Will you be able to buy self-driving cars with different profiles in there? Could you choose from "Scenic Route" to "late for work"? Could you even choose to have a discount if you listen to adds whilst being driven?

    I expect a rise in alcoholism when these things become mass-market, as you'll no longer need to worry about driving back from the pub!

    1. 142

      17mph unbraked, into the rear, is pretty significant. Most rear end collisions are much slower than that. (even if the car was initially going faster, it will have braked below that).

      17mph is like getting tackled from behind unexpectedly by a rugby player.

      It's gonna give whiplash, unless you were pre-braced, with your neck against the headrest.

      1. JeffyPoooh
        Pint

        "It's gonna give whiplash, unless you were pre-braced, with your neck against the headrest."

        Mercedes E-Class and likely others have spring loaded front seat headrests that will pop out on suitable crash detection command in order to reduce the whiplash effect, and thus reduce such injuries.

    2. d3vy

      Well we know that its a google car being tested so theres a good chance that they are employees of google... there is also a good chance that they were at work when this happened.

      If you were an employer anywhere and three of your employees were involved in an accident wouldn't you send them to hospital?

      Now think about it again if you were an employer in *America* land of the Pi claim and three of your employees were involved in an accident during work time whilst sitting inside one of your products and there are members of the public involved too wouldn't you INSIST that they got checked over?

      Add to the fact that being hit at 17mph is more than enough to cause whiplash its probably a very sensible precaution to get everyone checked out.

    3. Alan Brown Silver badge

      "If 3 people go to hospital in a rear-end-at-17mph collision, then the car isn't very crashworthy."

      The cars in question are Priuses and Lexus SUVs.

      17mph is a substantial hit. It means the driver behind probably didn't touch his brake until he was only about 4 metres behind the stopped car (normal road speed), if at all (slow moving traffic)

    4. Orv

      It's common practice in the U.S. to go to the hospital after just about any collision. If you turn out to have an injury and don't get it documented right away, the other driver's insurance might not pay, and you could get stuck with the bill. Health insurance will often shirk paying for things they think should have been covered by the other driver's liability insurance.

  15. David Lawrence

    Was the driver of the other car......

    ....distracted by using their mobile phone to update their status on FuckWitBook? "OMG guess what's in front of me?? Only one of them freakin Goog................" <message ends>

  16. FelixReg

    To avoid rear-enders

    Questions come to mind.

    Was this one of those itty-bitty Google cars? if so, welcome to the world of motorcycles. The big cars are out to get you. If they even see you. Which they often won't.

    You must defend yourself.

    How?

    Well, first, blink the brake lights until the guy behind responds.

    Second, even-out the crash. You can be moving forward slightly when he hits you. Always leave room to do so.

    Third, you may be able to dodge him by bailing to the gutter. That can backfire if he steers for the gutter to avoid you.

    Fourth, if you're a big car, let the guy behind you see the cars in front of you so he can see them stopping. Monitor whether he is watching them - or if he's just following your taillights. If the latter, lose him.

    Fifth, don't be invisible. Wobble in the lane. Vary your speed and acceleration. Create uncertainty in the other driver if you can't rely on him. Scare him if he's a bozo. He'll wake up quick. You want to do the opposite of how you lead a drunk driver to safety.

    1. Toltec

      Re: To avoid rear-enders

      In addition the car could warn the passengers of imminent impact so they can brace into the seat/headrest.

      The equivalent of noticing the driver look in the rear view mirror and and say "Oh sh!7" - true story...

      The auto-cars could also have active crumple zones, something like the ablative armour on tanks.

      1. d3vy

        Re: To avoid rear-enders

        I love the idea of the car shouting "oh shit" when it detects an imminent crash.

        It should also shout abuse at anyone who cuts it up or doesn't give way, "Oh you absolute c*ck end" for people failing to give way on a roundabout etc...

        Finally a prosthetic hand should come out of the dashboard and give the "Wanker" sign at anyone overtaking and then cutting in at a junction/closed lane on the mway.

    2. Six_Degrees

      Re: To avoid rear-enders

      You last suggestion, especially, has a name. It's called "reckless driving," and you can and will be ticketed for it.

      Behavior like that is also a big reason so many drivers think so many motorcyclists are assholes.

      1. FelixReg

        Re: To avoid rear-enders

        Matter of degree on that last suggestion.

        Your objectives are smooth and aware flow, while cutting the number of times cars do things like pull out right in front of you and kill you. Because if you're a bike (or, probably, a cute, unthreatening Google car) they *will* do that.

        Optimally, your pay-attention-to-me driving wants to wake up the sleeping guy who is just a circumstance away from "pulling out in front of you". At the same time, such driving wants to be effectively invisible to the aware drivers and those who present no threat. This is a subtle thing. Your "reckless" driving is not crazy. It's picking which side of the lane you're on at the moment. It's moving your head in a way that another driver will pick up unconsciously. In the case of you moving toward the car in back of you, you want just a moment of attention from that car. Enough to get the driver's foot off the gas and, maybe, on to the brake.

        I didn't mention that you can read and adjust for things like brand-of-car, driver body language, and exoskeleton (car) body language. Others have pointed out that you can sense when someone will be turning by their exoskeleton body language (wobble in lane, move to the turn-side of the lane, etc). That guy who plowed in to the Google car didn't look like a turner. That means he's less likely to slow down and more likely to want to hurry thru the intersection, depending on his age, sex, etc.

        And what about the other cars? Are any of them doing things likely to take that guy's eyes off your approaching bumper? Remember, when you're stopped, you are competing for the attention of the guy coming up from behind. If you want to operate at a high level, think about how you manage the guy behind him. You don't want the guy behind you to be rear-ended, either.

        Driver-less cars are probably acquiring this sort of knowledge. So a town filled with driver-less cars will feel a *lot* different from a human-driven town.

      2. Vic

        Re: To avoid rear-enders

        You last suggestion, especially, has a name. It's called "reckless driving,

        Where? 'Cos it certainly isn't in the UK - reckless driving has an entirely different definition.

        Vic.

  17. Missing Semicolon Silver badge
    Boffin

    Poor insurance risk

    According to insurance stats, people that get involved in accidents.... tend to get involved in accidents again.

    This is why your insurance premiums go up when you have an accident that is not your fault.

    One wonders if there is a particular driving style, or mode of thinking about driving, that causes other drivers to make mistakes to your detriment. And the Google car is exhibiting them.

    1. John Mangan

      Re: Poor insurance risk

      Somebody drove into my door panel in a car park at work. Fortunately s(he) was a decent human being and left a note with details.

      Please tell me how this makes it more likely that I will have another accident (while not in the car) and justifies the insurers increasing my premiums (I shit you not!).

      1. Oz

        Re: Poor insurance risk

        That's because the insurers choose to think that careless people use that car park so you are at risk of being hit again.

        1. John Mangan

          Re: Poor insurance risk

          Well I would like to see how they weight those statistics. I have been using that car park for at least eight years so that's at least 1600 parks (allowing for holiday/training/illness/etc.) with one (consequence-free) impact.

          But wait, they don't know what car park I use. That could have been the first and/or last time I used it. They don't know where I park and haven't asked.

          What a pity they don't conclude that it is used by (infrequently) careless but principled people so even if it happens again it won't cost them anything next time either!

          In fact their statistics give them no basis whatsoever for increasing my premium based on this one incident excepting providing a thinly veiled excuse.

          1. John Brown (no body) Silver badge

            Re: Poor insurance risk

            "increasing my premium...a thinly veiled excuse."

            It's an insurance company. Need we say more?

      2. d3vy

        Re: Poor insurance risk

        Why do your insurance company even know that happened?

        Youre not making a claim, there is no reason for you to get them involved - her insurance company would have dealt with the whole thing for you, My wife ran her two week old A1 into a mini a few months ago (I could have strangled her) it was a low speed impact partly the womans fault because she pulled out at a junction then stopped, but from an insurance point of view it was my wifes fault as she essentially hit a stationary car (she was checking that there was noting else coming in the opposite direction (there wasn't)

        Anyway, long story short, I called my insurance company they took our car away, gave the other person a courtesy car and took hers away, fixed them both and returned them.. the other parties insurance company were not involved at any point.

        1. heyrick Silver badge

          Re: Poor insurance risk

          "Why do your insurance company even know that happened?"

          Perhaps the small print that basically says that any damage caused to the vehicle that is not reported risks invalidating your insurance?

          I suspect the behaviour of insurance depends upon the country. We were involved in a minor crash - pulling out of a car park some inattentive person pulled out afterwards and smacked into the side. An accident form was filled in, an assessor came and looked at the car. The door was fixed and the bill sent to the other company. Our insurance didn't change as the positioning showed it was clearly not our fault.

          On the other hand, this country is a bloody nightmare for priority on the right, which is highly inconsistent and can change in a town by town basis so that a tractor can come out of a barely tarmac coated minor road on to a major road with a 90kph speed limit and the tractor has priority. Insane.

    2. Anonymous Coward
      Anonymous Coward

      Re: Poor insurance risk

      If Google's automated cars (other vendors are available) are as successful as hoped there should be fewer accidents which should lead to lower premiums. Where will that leave the car insurance industry?

      The manufacturers will own, or have best access to, the usage data. Would it be permissible to insure your vehicle with the manufacturer?

      Today the insurance giants trade on algorithms based around the common experiences of millions of people. Will their risk assessors be re-training to start judging software implementations instead?

      1. Alan Brown Silver badge

        Re: Poor insurance risk

        "Where will that leave the car insurance industry?"

        More to the point, with automated cars, fewer people will own cars so there are far fewer to insure (and roads will end up less congested too)

    3. Missing Semicolon Silver badge

      Re: Poor insurance risk

      Why the heck all the downvotes? Insurance companies' load your insurance after any claim, even if fully recovered from the third party. They do this for good, actuarial reasons, not just because they are bastards.

      Since this is a known statistical fact, and since the GoogleMobile seems to be a bit accident-prone, perhaps there's some similarity between its driving style and that of squishy drivers that keep getting rear-ended/pulled-out on etc.

      1. bristolmoose

        Re: Poor insurance risk

        Not sure this is true.

        Just renewed my wife's insurance and when getting the new insurer to call the old one to confirm no claim discount, they advised me that the old insurer had informed them that my wife had two non fault accidents to her name. I explained that these were accidents I had had and shouldn't be against her name, she wan't even in the car for either one. The new insurer said that they would have to include them for now and would take them off once I had sorted out the old insurer.

        The difference in premium? Absolutely nothing.

  18. W Donelson

    My young neighbour was hit...

    ... five years ago while stopped at a red traffic light, by a guy talking on his cellphone. He took full responsibility though, and had good liability insurance.

    She was badly injured, needs constant attention, and at only 24 now her life is ruined. My wife and I see ambulances outside her house 3-4 times a year, she is in constant pain, and recently her right side is going numb.

    The guy knows the continuing, probably life-long suffering he has caused, not a bad person at all, and suffers himself now.

    DON'T TALK AND DRIVE.

  19. The McV
    Stop

    Maintainance?

    The modern car comes with lots of electronic sensors and toys. After a couple of years in the very hostile (to electronics) environment of 'under the bonnet' - they start to irritatingly fail and glitch. Quite a lot seem to go 'tilt' - game over - and leave you stranded at the roadside. Now go to your friendly local garage to get it fixed at an exorbitant price. Will it get fixed properly? Do you really trust the grease monkey with delicate electronics?

    So now make the car drive itself....

    1 Do you trust that all the lines of code have been adequately tested?

    2 Do you trust the error recovery vectors in the code?

    3 Do you trust all the wiring, sensors & interconnections?

    4 Has it been hacked?

    Airlines use fly-by-wire & autoland etc, and that's just fine because the maintenance schedules are such that they get inspected on a regular basis by people who know what they're doing AND you have the flight crew monitoring.

    I see us getting railroaded into having the automated cars on the road, but I don't expect them to be safe until sometime next century......

    1. John Robson Silver badge

      Re: Maintainance?

      @The McV -I don't expect them to be safe until sometime next century......

      Which is still millenia ahead of people...

      They are going to be *safer than people*, which is the only sane first target, very soon indeed.

      Indeed for most purposes (in the climate of california) they probably already are...

    2. Six_Degrees

      Re: Maintainance?

      Autonomous cars don't have to be perfect. They just have to be better than cars driven by meatsacks.

      And that is a bar that's set incredibly low to begin with.

    3. Anonymous Coward
      Anonymous Coward

      Re: Maintainance?

      So now make the car drive itself....

      -so now let the human drive the car

      1 Do you trust that all the lines of code have been adequately tested?

      -how recently did it pass its test and how much of the Highway Code does it recollect?

      2 Do you trust the error recovery vectors in the code?

      -does it know what to do in an emergency such as loss of traction on a slippery surface?

      3 Do you trust all the wiring, sensors & interconnections?

      -is its brain fully functional and have its eyes been tested and corrected recently?

      4 Has it been hacked?

      -is it on drugs or just plain malicious?

      The biggest problem on the roads is people who think they are more competent than they really are.

    4. Alan Brown Silver badge

      Re: Maintainance?

      "After a couple of years in the very hostile (to electronics) environment of 'under the bonnet' - they start to irritatingly fail and glitch."

      My current daily driver is 12 years old, electroniced up to the hilt (including adaptive cruise control) and still as reliable as the day I bought it at about 9 months old (all problems have been mechanical other than the spark coilpacks all failing at about 10 years and the inevitable bulb failures.)

      Whatever you're driving (Renault perhaps? Something built by Lucas (Prince of darkness)?), perhaps you should consider avoiding that manufacturer.

      Making a reliable car is a known quantity. Making a car which has a constant procession of failures after 4-5 years is a planned way of extracting more money out of you.

  20. PsychicMonkey

    Not my fault guv...

    There seem to be a large number of comments here asking why the google cars (how many cars do they have in total?) have been in accidents in the 6 years they have been running. These comments appear to be saying that the automatic car must be at fault somehow.... I really can't see how that leap happens, if you rear end anything it is almost certainly always your fault. If you can't react fast enough to stop then you are too close at the speed you are going. Now I know this is an idealised view and I have been on both ends of this situation before, but the point still stands. The google car can not be at fault here, it had stopped as it should have done and as any fleshbag should have done, the person behind could not have been paying attention otherwise they would have seen the slowing car and done the same. I think it's fair to say that people will make far more errors on the road than an automated car will, but just as people always try to blame each other they are trying to blame the google car.

    1. Anonymous Coward
      Anonymous Coward

      Re: Not my fault guv...

      "I really can't see how that leap happens, if you rear end anything it is almost certainly always your fault."

      I had a very minor rear end seven years ago - I was waiting to turn right at a junction and Mrs. Futtwick went into me. Fortunately I saw her coming and released the brake, and damage was very minor. In my case, I needed a new number plate and that was it. Mitsubishis are like that.

      But what got me really angry was that she would not admit she was at fault. She simply wouldn't admit she had not been looking. As nobody was injured I couldn't call the police, but she then went off still believing her behaviour was nothing out of the ordinary.

      The problem with the Family Futtwick is that they obviously don't learn anything from experience.

      1. John Robson Silver badge

        Re: Not my fault guv...

        @AC - As nobody was injured I couldn't call the police

        Yes you can - property damage.

    2. Doctor Syntax Silver badge

      Re: Not my fault guv...

      "There seem to be a large number of comments here asking why the google cars (how many cars do they have in total?) have been in accidents in the 6 years they have been running."

      Those comments are largely in response to my original comment which started out with the word "If". I've no idea if the frequency is excessive. But if they are then it's reasonable to ask why.

      Much the same applies to stretches of roads with particularly high accident rates. An example would be one road near me. It has:

      - A cross-roads with another main road with particularly bad sight lines for both roads.

      - A junction on a bend where the minor road is a tangent to the main road. The main road has a 50 speed limit but the side road has a national speed-limit sign which catches headlights of drivers on the main road at night & looks as if it's the main road continuing in a straight line - even when you know it's a side road. I know there's been a head-on fatal collision on that bend.

      - Two more minor cross roads both with bad sight lines although one has little traffic & has been ameliorated with a mirror. Drivers on the main road approaching the latter are distracted by a flashing slow-down sign which would probably be triggered by a snail with a sufficiently large radar cross-section.

      - Some not very well cambered bends.

      It's plain that the major junction needs to be re-configured - the current staggered form just exposes crossing traffic for long than a straight crossing would do. It needs a roundabout or possibly a pair. The bend with the speed-limit needs reflectors to show the outline at night. The worst of the the other cross-roads needs traffic lights.

      What do they get? Signs advising drivers it's a high risk road and playing about with speed limits. You do not improve elevated risks on the road without asking why they're elevated and then tackling the specific issues. Don't blame the drivers if the road is the anomaly.

      In the case of Google cars IF there is an anomalous risk then one does need to ask why.

      1. Alan Brown Silver badge

        Re: Not my fault guv...

        "An example would be one road near me. It has:"

        County roading engineers who should be facing criminal charges.

    3. xeroks

      Re: Not my fault guv...

      being a tech website, I reckon a large proportion of commentards are predisposed to identify glitches and bugs, it's how we make our money, whether directly or indirectly.

      As a rule of thumb, if it's new, and not run in, there's probably something wrong with it somewhere, it's just a matter of finding the symptoms.

      For this - as others have said - without a control group, it's impossible to say whether the reported shunts are a indicator of a problem or not.

      If there is a problem with chocmobiles being shuntmagnets, it's likely more to do with the kind of signalling we all do inadvertently - road positioning, pre-turn wobbles, unneccessary braking etc. These give other road users information about us as drivers. If a googlecar gives contradictory signals, or none, then shunts may occur more often as people misread what the car's about to do.

  21. Zog_but_not_the_first
    Trollface

    It's a game

    And much more fun without the bendy pole on the back of the car and the chicken wire overhead..

  22. hatti

    Insurance claim

    Did the google podule put in a claim for e-whiplash?

  23. Paul Hovnanian Silver badge
    Terminator

    Bumper sticker

    Google cars need a bumper sticker: Driver is a T-1000. Back off!

    1. Destroy All Monsters Silver badge

      Re: Bumper sticker

      Why not a Confederate Flag to enrage liberals?

  24. sisk

    Yep, human at fault.

    This one's pretty clear cut. It actually looks remarkably similar to the time the college kid behind me had his eyes off the road and destroyed my motorcycle. Except, of course, that my motorcycle was being controlled by a human and didn't even have a computer on board, let alone one running the show.

  25. Kevin McMurtrie Silver badge
    Holmes

    Irony

    The company making self-driving cars happens to employ the largest pool of people who don't know how, and don't care how, to drive. It's amazing that swarms of motorcycle cops don't hang out there. I don't know if Google and LinkedIn make "donations" to the police, but 50 traffic violations per minute per intersection sounds like a lot of money.

  26. Red Bren
    Terminator

    As this is Google...

    Are the Google cars recording every interaction they have with other vehicles? Every failure to indicate, every red light jumped, etc? How long before you can google the registration of the arsehole who cut you up this morning, find out where they tend to be a given time and go offer them some helpful driving advice? How long before Google cars will adapt their driving style based on the history they have collected for the registration plates of the surrounding vehicles? How long before they start taking revenge...

  27. iLuddite

    the tech will be fine

    The EULA will be the killer.

POST COMMENT House rules

Not a member of The Register? Create a new account here.

  • Enter your comment

  • Add an icon

Anonymous cowards cannot choose their icon

Biting the hand that feeds IT © 1998–2021