Another fatal Tesla crash
Discussion
Mr2Mike said:
DJP31 said:
The only ones I’ve seen are the two posted above, which aren’t exactly great!
Check the link I posted. DJP31 said:
The braking isn’t exactly autonomous, it’s automatic emergency braking to try and minimise, not prevent, an impact. I imagine the reason it’s not part of the autopilot set up is due to the reasons given by Max_Torque above - i.e. cost and complexity.
That would be fine, certainly better than nothing. I don't believe such a system would add significantly to the cost of an autonomous car that already has all the sensors in place. Someone posted above that Tesla are claiming to have the feature anyway, so why doesn't it work?Edited by Mr2Mike on Tuesday 3rd April 14:29
The stationary one is very good, and I wonder if the speed of the vehicle determines whether it stops before it hits or not. In my Tesla the collision warning comes on in plenty of time to stop (so far anyway!), but the auto emergency braking cuts in if no braking is detected and the car looks like an impact is inevitable. Whether that’s speed dependent I don’t know.
One thing for sure though, the technology is still very much work in progress!
Test drove a Model S for a couple of days after Xmas and discovered that the AP system doesn't really like hump back bridges or even mild rises in the road on a slight bend.
First incident - good, clear lane markings on a dual carriageway and road rises over maybe 50 metres before dropping away and slightly to the right. AP had the car positioned nicely in lane 2 with no traffic around for a good 50m in any direction. Crested the rise and the car violently went to the left i guess to try to follow the line it could see, i caught it just as it moved but
Second incident - hump back bridge on opposite side of dual carriageway. Again, crested the peak of the bridge and the car started to dive left but expecting it this time i just gave it a quick tug to the right to pull the car back.
I mentioned this the sales guy when i dropped it off and he confirmed that the camera sensors can lose the sign and then it searches for nearest object to lock on to. He did say that as the database of Tesla journeys build up for that stretch of road (car continuously monitors GPS position, driver feedback etc which gets put back into the system to help "learn" roads) the system should then stay in the correct lane and not dive to one side.
My Model S arrives next month so i'll see if it's learnt anything or whether I shall have to teach it
First incident - good, clear lane markings on a dual carriageway and road rises over maybe 50 metres before dropping away and slightly to the right. AP had the car positioned nicely in lane 2 with no traffic around for a good 50m in any direction. Crested the rise and the car violently went to the left i guess to try to follow the line it could see, i caught it just as it moved but
Second incident - hump back bridge on opposite side of dual carriageway. Again, crested the peak of the bridge and the car started to dive left but expecting it this time i just gave it a quick tug to the right to pull the car back.
I mentioned this the sales guy when i dropped it off and he confirmed that the camera sensors can lose the sign and then it searches for nearest object to lock on to. He did say that as the database of Tesla journeys build up for that stretch of road (car continuously monitors GPS position, driver feedback etc which gets put back into the system to help "learn" roads) the system should then stay in the correct lane and not dive to one side.
My Model S arrives next month so i'll see if it's learnt anything or whether I shall have to teach it
arfursleep said:
Test drove a Model S for a couple of days after Xmas and discovered that the AP system doesn't really like hump back bridges or even mild rises in the road on a slight bend.
First incident - good, clear lane markings on a dual carriageway and road rises over maybe 50 metres before dropping away and slightly to the right. AP had the car positioned nicely in lane 2 with no traffic around for a good 50m in any direction. Crested the rise and the car violently went to the left i guess to try to follow the line it could see, i caught it just as it moved but
Second incident - hump back bridge on opposite side of dual carriageway. Again, crested the peak of the bridge and the car started to dive left but expecting it this time i just gave it a quick tug to the right to pull the car back.
I mentioned this the sales guy when i dropped it off and he confirmed that the camera sensors can lose the sign and then it searches for nearest object to lock on to. He did say that as the database of Tesla journeys build up for that stretch of road (car continuously monitors GPS position, driver feedback etc which gets put back into the system to help "learn" roads) the system should then stay in the correct lane and not dive to one side.
My Model S arrives next month so i'll see if it's learnt anything or whether I shall have to teach it
The most recent software update has made a big difference and the car is much less prone to this behaviour than before. I haven't had any problems myself, and it happily dealt with a crest on the A31 heading towards Ringwood which in the past it hasn't.First incident - good, clear lane markings on a dual carriageway and road rises over maybe 50 metres before dropping away and slightly to the right. AP had the car positioned nicely in lane 2 with no traffic around for a good 50m in any direction. Crested the rise and the car violently went to the left i guess to try to follow the line it could see, i caught it just as it moved but
Second incident - hump back bridge on opposite side of dual carriageway. Again, crested the peak of the bridge and the car started to dive left but expecting it this time i just gave it a quick tug to the right to pull the car back.
I mentioned this the sales guy when i dropped it off and he confirmed that the camera sensors can lose the sign and then it searches for nearest object to lock on to. He did say that as the database of Tesla journeys build up for that stretch of road (car continuously monitors GPS position, driver feedback etc which gets put back into the system to help "learn" roads) the system should then stay in the correct lane and not dive to one side.
My Model S arrives next month so i'll see if it's learnt anything or whether I shall have to teach it
I think you'll be impressed with the improvement, but don't be arfursleep when you use it!
hot metal said:
Why would anybody trust their lives with this system is beyond me, stop being so lazy & drive the f**king things.
Elon Musk says Tesla`s system is better than a person, a certain type of person maybe but not all .
Elon Musk says Tesla`s system is better than a person, a certain type of person maybe but not all .
A fool would trust their lives with it, and some have with a couple paying the ultimate price. It’s got nothing to do with being lazy either, that’s a facile comment. It is a driver aid and used in the right circumstances and with the driver paying attention it is extremely good.
I’ve been from Surrey to Cardiff and back over the last two days, M25 and M4. 90% of those miles were on autopilot and it was faultless. Until you’ve tried it properly you don’t appreciate how much mental energy we use making micro adjustments to steering and throttle pressure. Being relieved of those results in being much fresher at the end of the journey, and also means I can have higher concentration levels on what’s going on around me.
What Musk means, as I expect you understand really, is that technology is better than a person when used within its limitations, and that’s probably true of most technology. It’s the human that has to understand the limitations and work within them.
DJP31 said:
A fool would trust their lives with it, and some have with a couple paying the ultimate price. It’s got nothing to do with being lazy either, that’s a facile comment. It is a driver aid and used in the right circumstances and with the driver paying attention it is extremely good.
I’ve been from Surrey to Cardiff and back over the last two days, M25 and M4. 90% of those miles were on autopilot and it was faultless. Until you’ve tried it properly you don’t appreciate how much mental energy we use making micro adjustments to steering and throttle pressure. Being relieved of those results in being much fresher at the end of the journey, and also means I can have higher concentration levels on what’s going on around me.
What Musk means, as I expect you understand really, is that technology is better than a person when used within its limitations, and that’s probably true of most technology. It’s the human that has to understand the limitations and work within them.
hot metal said:
Don`t agree with any of this, more input from the driver heightens attention levels, keeps a driver focused. To many driver aids just promotes sloppiness
imho.
ABS makes drivers think theirs cars won’t crash, maybe you think that should be banned too? It all comes down to the central point, and that is the human element. Used properly, which most of the time they are, they assist the driver.imho.
REALIST123 said:
DJP31 said:
A fool would trust their lives with it, and some have with a couple paying the ultimate price. It’s got nothing to do with being lazy either, that’s a facile comment. It is a driver aid and used in the right circumstances and with the driver paying attention it is extremely good.
I’ve been from Surrey to Cardiff and back over the last two days, M25 and M4. 90% of those miles were on autopilot and it was faultless. Until you’ve tried it properly you don’t appreciate how much mental energy we use making micro adjustments to steering and throttle pressure. Being relieved of those results in being much fresher at the end of the journey, and also means I can have higher concentration levels on what’s going on around me.
What Musk means, as I expect you understand really, is that technology is better than a person when used within its limitations, and that’s probably true of most technology. It’s the human that has to understand the limitations and work within them.
hot metal said:
Don`t agree with any of this, more input from the driver heightens attention levels, keeps a driver focused. To many driver aids just promotes sloppiness
imho.
Not sure I completely agree. I like the adaptive cruise control. It means I have more "bandwidth" to consider the actions of more road-users in greater detail, as I need to focus less on the car immediately ahead of me. One should always do both but it does enable a subtle shift of attention that creates additional safety. I would agree if one uses the extra bandwidth to play with the sat nav or contemplate one's testicles, it is a detriment rather than an improvement.imho.
AstonZagato said:
Not sure I completely agree. I like the adaptive cruise control. It means I have more "bandwidth" to consider the actions of more road-users in greater detail, as I need to focus less on the car immediately ahead of me. One should always do both but it does enable a subtle shift of attention that creates additional safety. I would agree if one uses the extra bandwidth to play with the sat nav or contemplate one's testicles, it is a detriment rather than an improvement.
My experience with Auto Pilot is similar. It is a tool that helps me as a driver, it is not a replacement for my attention or my skills and judgement.I suspect good drivers will be better with such systems because it allows some of their concentration to be moved into other areas of driving, and bad drivers will be worse because it allows them not to concentrate at all. Whether it's better or worse overall, I think remains to be seen.
kambites said:
I suspect good drivers will be better with such systems because it allows some of their concentration to be moved into other areas of driving, and bad drivers will be worse because it allows them not to concentrate at all. Whether it's better or worse overall, I think remains to be seen.
I think this is a fair comment , Gassing Station | EV and Alternative Fuels | Top of Page | What's New | My Stuff