VW Adaptive Cruise/Anti-Collision Radar - Nearly crashed!

[DonkeyApple]

Crusoe said:

Seen some triggered by wind blown debris and heard a few mention it on here. Foil crisp bag at the wrong height blown past the sensor and it will think you are about to kill a child. Most of these systems are fine if there is good traction on the road surface and there is a decent gap to the car behind. Some have reduced the speed that the full braking and avoidance will kick in to under 30mph so there are less false occurrences but I still wouldn't want to drive one on a wintery day with low traction around a busy town or city.

The system is clearly very good 99% of the time and a beneficial development.

But would people tolerate an airbag system that was only 99% perfect and ran a credible chance of going off for absolutely no reason?

It strikes me that this system highlights what is only going to be a growing issue which is human drivers sharing the road with computer drivers. The two think and react very differently and either can do so for the better or for worse it seems.

[Sheepshanks]

boz1 said:

Sheepshanks said:

The message is that you need to adapt your driving to be aware that the car in front might inexplicably do a full-on emergency stop.

At least, I assume you're not being serious, right?

This is precisely my point. Nobody actually drives like that. Even those who think they do, probably don't.

If you look at the posts immediately after I posted my comment you'll find that several PHers assert they do always drive like that.

[TwigtheWonderkid]

Sheepshanks said:

boz1 said:

Sheepshanks said:

The message is that you need to adapt your driving to be aware that the car in front might inexplicably do a full-on emergency stop.

At least, I assume you're not being serious, right?

This is precisely my point. Nobody actually drives like that. Even those who think they do, probably don't.

If you look at the posts immediately after I posted my comment you'll find that several PHers assert they do always drive like that.

Everybody should always leave a big enough gap from the car in front so as not to hit them should they slam on the brakes, be it for good reason or for no reason. That's obvious. The fact that we don't always do that all of the time is our responsibility.

[Kawasicki]

I'd like to try and clear some stuff up.

You see a car in front of you start to turn off. You see the road in front of this car is clear. You determine that it is highly unlikely that the car will stop. You accelerate towards a car that is still in your path, because that is how your brain works, using risk assessment and predictions based on experience.

The computer sees a moving target in front of the car. It doesn't have any intelligence and very poor vision. It doesn't understand that the car is leaving your path and that it is highly unlikely it will change course. It just sees a lump of metal that is in your path and it slams on the brakes.

Get used to it. There will only be more of this. As much as I welcome the technology, it is way behind the marketing at the moment. A computer will only be able to drive as well as a good driver when it is as intelligent as a good driver. When computers are as intelligent as a good driver, we will have bigger issues to worry about. As for poor drivers, well they are much easier to better. Some sort of looking out of the car detection would be helpful.

[Trustmeimadoctor]

ClockworkCupcake said:

I don't find that at all. I'm looking as far ahead as the radar is and judging closing speeds, and am generally indicating to change lanes before the system even starts to slow down.

If you're being slowed down when coming up behind a truck, then I suggest you're not thinking ahead / looking ahead enough.

I will video it

[DonkeyApple]

TwigtheWonderkid said:

Sheepshanks said:

boz1 said:

Sheepshanks said:

The message is that you need to adapt your driving to be aware that the car in front might inexplicably do a full-on emergency stop.

At least, I assume you're not being serious, right?

This is precisely my point. Nobody actually drives like that. Even those who think they do, probably don't.

If you look at the posts immediately after I posted my comment you'll find that several PHers assert they do always drive like that.

Everybody should always leave a big enough gap from the car in front so as not to hit them should they slam on the brakes, be it for good reason or for no reason. That's obvious. The fact that we don't always do that all of the time is our responsibility.

Just to remind you: You aren't correct.

[NDNDNDND]

An interesting comment regarding autonomous cars cropped up on Jalopnik a while ago. They were discussing the number of google car 'interventions' (i.e. when the human driver has to override the car) and discussing the crash rate compared to human drivers. Someone said the google car only has to be better than the average for it be beneficial. Someone pointed out that being 'better than average' will make driving more dangerous for just under the top 50% of drivers.

This automatic braking system seems to be much like that. It makes incompetent or unobservant drivers safer, but becomes a patronising and potentially dangerous liability for those who actually know what they're doing.

I think insurers are reducing the risks associated with cars fitted with the system, though. And I'd guess they're the ones with the statistics...

[ClockworkCupcake]

Trustmeimadoctor said:

ClockworkCupcake said:

I don't find that at all. I'm looking as far ahead as the radar is and judging closing speeds, and am generally indicating to change lanes before the system even starts to slow down.

If you're being slowed down when coming up behind a truck, then I suggest you're not thinking ahead / looking ahead enough.

I will video it

Just to clarify, are we talking about the speed actually dropping here, or are we talking about when the display flashes up a picture to say the system as identified a vehicle in front of you?

I experience the latter but I don't experience the former as I am already changing lanes before the system starts to try to match the speed of the car in front. I have found that you need to be smoothly changing lanes in good anticipation, rather than bombing up behind the car in front and then pulling out sharpish like a lot of drivers do, so perhaps it's a driving style thing. Having said that, I am changing lanes maybe slightly earlier than I would otherwise so perhaps I am driving round the issue (no pun intended).

As with so much of these things, you need to work with the driver aid rather than against it.

Edited by ClockworkCupcake on Tuesday 26th January 14:02

[daniel1920]

Kawasicki said:

I'd like to try and clear some stuff up.

You see a car in front of you start to turn off. You see the road in front of this car is clear. You determine that it is highly unlikely that the car will stop. You accelerate towards a car that is still in your path, because that is how your brain works, using risk assessment and predictions based on experience.

No, you slow down and wait and do not overtake at a junction. H2H

[mybrainhurts]

daniel1920 said:

Kawasicki said:

I'd like to try and clear some stuff up.

You see a car in front of you start to turn off. You see the road in front of this car is clear. You determine that it is highly unlikely that the car will stop. You accelerate towards a car that is still in your path, because that is how your brain works, using risk assessment and predictions based on experience.

No, you slow down and wait and do not overtake at a junction. H2H

If there's nothing else at the junction and your road ahead is clear, what's wrong with pulling out to go past? You might as well say overtaking is out of order.

[daniel1920]

mybrainhurts said:

daniel1920 said:

Kawasicki said:

I'd like to try and clear some stuff up.

You see a car in front of you start to turn off. You see the road in front of this car is clear. You determine that it is highly unlikely that the car will stop. You accelerate towards a car that is still in your path, because that is how your brain works, using risk assessment and predictions based on experience.

No, you slow down and wait and do not overtake at a junction. H2H

If there's nothing else at the junction and your road ahead is clear, what's wrong with pulling out to go past? You might as well say overtaking is out of order.

https://dl.weeaboo.com/dsa/hwc/highway-code/rule-167.html

Seen a handful of threads on here where someone explains there crash, and how the junction was perfectly clear, then be at fault and not have insurance pay out.
I agree sometimes it seems OTT but if you go for a blanket do not do it, this site would be about 10 threads lighter, including this one...

[DonkeyApple]

mybrainhurts said:

daniel1920 said:

Kawasicki said:

I'd like to try and clear some stuff up.

You see a car in front of you start to turn off. You see the road in front of this car is clear. You determine that it is highly unlikely that the car will stop. You accelerate towards a car that is still in your path, because that is how your brain works, using risk assessment and predictions based on experience.

No, you slow down and wait and do not overtake at a junction. H2H

If there's nothing else at the junction and your road ahead is clear, what's wrong with pulling out to go past? You might as well say overtaking is out of order.

It's not something that I would rush to do though because when the idiot cyclist who pulled out thinking it was clear rolls over my bonnet with the surprised look of a retard in a thunderstorm I would rather this natural selection event occurred on my side of the road.

[Trustmeimadoctor]

ClockworkCupcake said:

Just to clarify, are we talking about the speed actually dropping here, or are we talking about when the display flashes up a picture to say the system as identified a vehicle in front of you?

I experience the latter but I don't experience the former as I am already changing lanes before the system starts to try to match the speed of the car in front. I have found that you need to be smoothly changing lanes in good anticipation, rather than bombing up behind the car in front and then pulling out sharpish like a lot of drivers do, so perhaps it's a driving style thing. Having said that, I am changing lanes maybe slightly earlier than I would otherwise so perhaps I am driving round the issue (no pun intended).

As with so much of these things, you need to work with the driver aid rather than against it.

Edited by ClockworkCupcake on Tuesday 26th January 14:02

Speed dropping I'm learning slowly what distance it likes to keep

[mybrainhurts]

daniel1920 said:

mybrainhurts said:

daniel1920 said:

Kawasicki said:

I'd like to try and clear some stuff up.

You see a car in front of you start to turn off. You see the road in front of this car is clear. You determine that it is highly unlikely that the car will stop. You accelerate towards a car that is still in your path, because that is how your brain works, using risk assessment and predictions based on experience.

No, you slow down and wait and do not overtake at a junction. H2H

If there's nothing else at the junction and your road ahead is clear, what's wrong with pulling out to go past? You might as well say overtaking is out of order.

https://dl.weeaboo.com/dsa/hwc/highway-code/rule-167.html

Seen a handful of threads on here where someone explains there crash, and how the junction was perfectly clear, then be at fault and not have insurance pay out.
I agree sometimes it seems OTT but if you go for a blanket do not do it, this site would be about 10 threads lighter, including this one...

Applicable only if your view of the junction is in doubt. That's a different matter.

[daniel1920]

mybrainhurts said:

Applicable only if your view of the junction is in doubt. That's a different matter.

100% the car infront is obscuring your view. A small child is shorter than a bonnet

[DonkeyApple]

mybrainhurts said:

daniel1920 said:

mybrainhurts said:

daniel1920 said:

Kawasicki said:

I'd like to try and clear some stuff up.

You see a car in front of you start to turn off. You see the road in front of this car is clear. You determine that it is highly unlikely that the car will stop. You accelerate towards a car that is still in your path, because that is how your brain works, using risk assessment and predictions based on experience.

No, you slow down and wait and do not overtake at a junction. H2H

If there's nothing else at the junction and your road ahead is clear, what's wrong with pulling out to go past? You might as well say overtaking is out of order.

https://dl.weeaboo.com/dsa/hwc/highway-code/rule-167.html

Seen a handful of threads on here where someone explains there crash, and how the junction was perfectly clear, then be at fault and not have insurance pay out.
I agree sometimes it seems OTT but if you go for a blanket do not do it, this site would be about 10 threads lighter, including this one...

Applicable only if your view of the junction is in doubt. That's a different matter.

I would think that it would be impossible not to be unless the vehicle in front were totally transparent? Any such action would contain an element of risk as everytime part of your view is going to be obscured?

[boz1]

Sheepshanks said:

If you look at the posts immediately after I posted my comment you'll find that several PHers assert they do always drive like that.

Yes, I know they do. I don't believe them. Indeed, I expect a lot of people fail to realise just how quickly a modern car will stop if you apply full braking force.

I'm just being honest. If I look around the roads, I see plenty of people travelling closer than me and very few leaving noticeably bigger braking gaps. I still consider that, realistically, there is a chance I would hit someone who did a full emergency stop for no reason.

Think about whenever you may have had to perform an emergency stop. I would be surprised if hardly any of us has had another driver perform a full emergency stop for no reason. Normally, you will also observe and have time to react to the situation that might cause the driver in front to do so (e.g. a crash occurs ahead, a pedestrian steps out, etc.).

If you don't see such an event, then you're not expecting an emergency stop and your first warning will be the brake lights of emergency-stopper, then the time (and therefore distance) that you require to react to the braking, realise how hard they are braking and react accordingly will be that much greater.

According to Autoexpress, an Insignia will pull up from 30mph in 8.7m. http://www.autoexpress.co.uk/car-reviews/38948/vau...

30mph is over 13 metres per second. Let's say you are in an Insignia and are following another Insignia at 30 mph with a reasonable-feeling three car lengths gap. I'm not saying that's enough, I'm just saying it's pretty normal; the "two second rule" would call for six car lengths at this speed. Three Insignia lengths is a little over 13 metres.

The front Insignia slams on the anchors, which means its going to stop dead about 22 metres from the position you were when at the instant he started braking. You then have at most 1 second to start braking at full force; you can brake at less than full force only if you start within that second. Otherwise, you are going to hit him.

Sure it might not be a very serious accident, but you will hit him. Of course, most emergency braking situations don't happen like this because, consciously or otherwise, you'll have more reaction time because you'll see the incident that causes the braking. But this is what the "random emergency stop" scenario is like.

[SturdyHSV]

I've read most of the thread and skimmed the latter half but one thing that may merit considering is whether the OP nailing the throttle pedal ( this is an assumption ) may have been contributed to it?

I'm only suggesting it because people do hit the wrong pedals, so perhaps although the incident may have just merited a warning from the system in normal conditions, the OP giving one of the pedals a good jabbing may have triggered the 'driver wants to stop but is panicked' mechanism?

I'm not sure if such a mechanism is programmed in, but it could easily be touted as a safety feature to help protect those of a more ungulate nature

[ClockworkCupcake]

Trustmeimadoctor said:

Speed dropping I'm learning slowly what distance it likes to keep

Yes, I have found that I have to change lanes a little sooner than I would have otherwise.

[mybrainhurts]

DonkeyApple said:

mybrainhurts said:

daniel1920 said:

mybrainhurts said:

daniel1920 said:

Kawasicki said:

I'd like to try and clear some stuff up.

You see a car in front of you start to turn off. You see the road in front of this car is clear. You determine that it is highly unlikely that the car will stop. You accelerate towards a car that is still in your path, because that is how your brain works, using risk assessment and predictions based on experience.

No, you slow down and wait and do not overtake at a junction. H2H

If there's nothing else at the junction and your road ahead is clear, what's wrong with pulling out to go past? You might as well say overtaking is out of order.

https://dl.weeaboo.com/dsa/hwc/highway-code/rule-167.html

Seen a handful of threads on here where someone explains there crash, and how the junction was perfectly clear, then be at fault and not have insurance pay out.
I agree sometimes it seems OTT but if you go for a blanket do not do it, this site would be about 10 threads lighter, including this one...

Applicable only if your view of the junction is in doubt. That's a different matter.

I would think that it would be impossible not to be unless the vehicle in front were totally transparent? Any such action would contain an element of risk as everytime part of your view is going to be obscured?

No, it's not impossible. If there's the slightest doubt, I don't proceed, but there are plenty of junctions that can be clearly assessed as clear. It's bloody annoying when the car in front proceeds as you suggest in such circumstances and unnecessarily inhibits progress.

I drive between 50,000 and 80,000 miles a year and have never been caught out in error, so not impossible.