This topic has been archived, and won't accept reply postings.
Sounds like a dodgy mix to me
https://www.bicycling.com/culture/cyclist-killed-by-tesla-car-with-self-dri...
https://www.bicycling.com/culture/cyclist-killed-by-tesla-car-with-self-dri...
5
In reply to LeeWood:
The article says that it's not known whether the car was in self-driving mode or not at the time of the accident. The driver could have been texting, asleep, drunk, half blind or distracted.
The article says that it's not known whether the car was in self-driving mode or not at the time of the accident. The driver could have been texting, asleep, drunk, half blind or distracted.
10
1
In reply to Rigid Raider:
For completeness due to lack of evidence, that assumption could also apply to the cyclist couldn't it in all fairness?
> The article says that it's not known whether the car was in self-driving mode or not at the time of the accident. The driver could have been texting, asleep, drunk, half blind or distracted.
For completeness due to lack of evidence, that assumption could also apply to the cyclist couldn't it in all fairness?
7
2
In reply to LeeWood:
Its sure there are a few unknowns here, but the basic issue is - can self-drive and cyclists mix on the road with safety?
Given that humans are poor at spotting cyclists, and they are the ones which program these cars it leaves me uneasy.
I wasn't aware that such vehicles were even loose on the roads now, but imagine - you're at a junction with right of way and believe the driver has seen you ... and move out. NO - given the sightest evidence that its a self-drive I would get right off the road ! Perhaps they should be fitted with green flashing beacons !!
Its sure there are a few unknowns here, but the basic issue is - can self-drive and cyclists mix on the road with safety?
Given that humans are poor at spotting cyclists, and they are the ones which program these cars it leaves me uneasy.
I wasn't aware that such vehicles were even loose on the roads now, but imagine - you're at a junction with right of way and believe the driver has seen you ... and move out. NO - given the sightest evidence that its a self-drive I would get right off the road ! Perhaps they should be fitted with green flashing beacons !!
2
12
In reply to paul_in_cumbria:
is that likely if someone has reached 80 as keen and active cyclist? A moment's consideration makes it an unsafe and unequal comparison
is that likely if someone has reached 80 as keen and active cyclist? A moment's consideration makes it an unsafe and unequal comparison
3
2
In reply to LeeWood:
My money would be on self-drive cars can see cyclists better than meat-bags. They don't get distracted for starters, and they have multiple sensors, and quicker reaction times. They follow the rules of the roads (mostly), and won't be reckless because they are late getting somewhere.
My money would be on self-drive cars can see cyclists better than meat-bags. They don't get distracted for starters, and they have multiple sensors, and quicker reaction times. They follow the rules of the roads (mostly), and won't be reckless because they are late getting somewhere.
Post edited at 10:29
34
In reply to JoshOvki:
I read an article a couple of years suggesting that not only can the autonomous car technology see cyclists, it could actually anticipate movement from the cyclist based on subtle changes in body language that a human would be hard pressed to see.
If the car involved here was a Vauxhall Corsa i doubt it would have been mentioned but obviously mentioning Tesla creates an inherent ‘killer robot cars on the loose’ feeling which might* be completely unjustified.
*I’m not giving Tesla a free pass here, i jus front think these kind of stories are helpful until the facts are know. The recent case in Florida where the car failed to spot a turning lorry and the driver was killed rightly flagged up some of the limitation in the current implementation of the technology and reminded people not to get too Carrie away with what they think the car can safely do.
I read an article a couple of years suggesting that not only can the autonomous car technology see cyclists, it could actually anticipate movement from the cyclist based on subtle changes in body language that a human would be hard pressed to see.
If the car involved here was a Vauxhall Corsa i doubt it would have been mentioned but obviously mentioning Tesla creates an inherent ‘killer robot cars on the loose’ feeling which might* be completely unjustified.
*I’m not giving Tesla a free pass here, i jus front think these kind of stories are helpful until the facts are know. The recent case in Florida where the car failed to spot a turning lorry and the driver was killed rightly flagged up some of the limitation in the current implementation of the technology and reminded people not to get too Carrie away with what they think the car can safely do.
8
In reply to JoshOvki:
Ironically this seems to be causing problems. It looks like self driving prototypes have an excessive number of low speed crashes where the other driver is at fault - meat bags operate away from the rules and have a baked in assumption that other cars will.
> They follow the rules of the roads (mostly)
Ironically this seems to be causing problems. It looks like self driving prototypes have an excessive number of low speed crashes where the other driver is at fault - meat bags operate away from the rules and have a baked in assumption that other cars will.
1
In reply to LeeWood:
It's my belief that Britain's narrow, twisty, congested roads based on illogical meandering packhorse trails and 18th century turnpikes will defeat self-driving cars. However smart motorways are here, most cars already have the ability to follow the car in front by radar and it's only a matter of time before smart motorways are used for their proper purpose, which will be for platoons of cars a couple of feet apart, all moving at the same speed, which will double or treble the capacity of the busiest roads. You can see how platooning will work here: https://peloton-tech.com/
All the technology exists; it just needs to be joined up.
It's my belief that Britain's narrow, twisty, congested roads based on illogical meandering packhorse trails and 18th century turnpikes will defeat self-driving cars. However smart motorways are here, most cars already have the ability to follow the car in front by radar and it's only a matter of time before smart motorways are used for their proper purpose, which will be for platoons of cars a couple of feet apart, all moving at the same speed, which will double or treble the capacity of the busiest roads. You can see how platooning will work here: https://peloton-tech.com/
All the technology exists; it just needs to be joined up.
4
3
In reply to LeeWood:
I heard a story that an autonomous car in the UK refused to move off at a junction where it had right of way because a cyclist on the give-way road was doing a track stand and it interpreted it as a suicidal nutter constantly about to dash forward.
The point of the story was that you do need to do a fair bit of localisation for situations and habits it won’t have seen in its US test facility, but they do try and fail safely.
I heard a story that an autonomous car in the UK refused to move off at a junction where it had right of way because a cyclist on the give-way road was doing a track stand and it interpreted it as a suicidal nutter constantly about to dash forward.
The point of the story was that you do need to do a fair bit of localisation for situations and habits it won’t have seen in its US test facility, but they do try and fail safely.
3
In reply to LeeWood:
Yes.
I don't think programming - in this context - works how you think it does.
> Its sure there are a few unknowns here, but the basic issue is - can self-drive and cyclists mix on the road with safety?
Yes.
> Given that humans are poor at spotting cyclists, and they are the ones which program these cars it leaves me uneasy.
I don't think programming - in this context - works how you think it does.
12
In reply to wintertree:
Yeah! I think they have changed it so it doesn't do things exactly text book now, although I am not 100% on that.
When I was doing my CBT the instructor said something I hadn't thought of before:
"If everyone followed the highway code to the text, the only crashes would be mechanical failure", turns out he was pretty much spot on (apart from a week later I crashed my bike on a cattle grid and broke my arm, it wasn't covered by the highway code...).
Yeah! I think they have changed it so it doesn't do things exactly text book now, although I am not 100% on that.
When I was doing my CBT the instructor said something I hadn't thought of before:
"If everyone followed the highway code to the text, the only crashes would be mechanical failure", turns out he was pretty much spot on (apart from a week later I crashed my bike on a cattle grid and broke my arm, it wasn't covered by the highway code...).
1
In reply to LeeWood:
They're proving safer than human drivers so far in tests, given it's early days I think that bodes well for the future. Whether they'll prove as proactive and intuitive as a good, experienced and self-aware driver can be I'm dubious but that's not who is likely to knock you off and even they have lapses of attention.
Sort of misses the point about machine learning and the reasons *why* motorists are poor at seeing and avoiding cyclists. We are poor at seeing cyclists but it has more to do with how our vision works in the moment, how what we see is processed and the things the brain is hardwired to pay attention to than the value we as society (or engineers/developers) will assign to cyclists. That and the fact we're not taught as drivers strategies to deal with our deficiencies, real deficiencies many simply aren't aware of or willing to accept they have.
Also machine vision can be multispectral which improves significantly on human vision in visually cluttered dangerous environments like dark wet cityscapes full of illuminated signs, signals and streetlights. There's also the possibility of vehicles networking to share hazard perception information which is basically impossible for humans to do (ok, hazard lights but they're crap).
I'd suggest we don't panic yet.
jk
> Its sure there are a few unknowns here, but the basic issue is - can self-drive and cyclists mix on the road with safety?
They're proving safer than human drivers so far in tests, given it's early days I think that bodes well for the future. Whether they'll prove as proactive and intuitive as a good, experienced and self-aware driver can be I'm dubious but that's not who is likely to knock you off and even they have lapses of attention.
> Given that humans are poor at spotting cyclists, and they are the ones which program these cars it leaves me uneasy.
Sort of misses the point about machine learning and the reasons *why* motorists are poor at seeing and avoiding cyclists. We are poor at seeing cyclists but it has more to do with how our vision works in the moment, how what we see is processed and the things the brain is hardwired to pay attention to than the value we as society (or engineers/developers) will assign to cyclists. That and the fact we're not taught as drivers strategies to deal with our deficiencies, real deficiencies many simply aren't aware of or willing to accept they have.
Also machine vision can be multispectral which improves significantly on human vision in visually cluttered dangerous environments like dark wet cityscapes full of illuminated signs, signals and streetlights. There's also the possibility of vehicles networking to share hazard perception information which is basically impossible for humans to do (ok, hazard lights but they're crap).
> I wasn't aware that such vehicles were even loose on the roads now, but imagine - you're at a junction with right of way and believe the driver has seen you ... and move out. NO - given the sightest evidence that its a self-drive I would get right off the road ! Perhaps they should be fitted with green flashing beacons !!
I'd suggest we don't panic yet.
jk
Post edited at 11:34
2
In reply to yorkshireman:
It really depends on the self-driving system in question, as there's a wide range of capabilities in the works.
Waymo's (aka Google) system is probably the most advanced at the moment and likely has some of the capabilities you describe. I believe Tesla's system is less advanced though, in particular they have a less capable sensor array compared to the waymo cars. (Id need to double check, but I think it's the LIDAR sensor that the tesla cars are missing). There will also be a lot of differences in the software of each car.
> I read an article a couple of years suggesting that not only can the autonomous car technology see cyclists, it could actually anticipate movement from the cyclist based on subtle changes in body language that a human would be hard pressed to see.
It really depends on the self-driving system in question, as there's a wide range of capabilities in the works.
Waymo's (aka Google) system is probably the most advanced at the moment and likely has some of the capabilities you describe. I believe Tesla's system is less advanced though, in particular they have a less capable sensor array compared to the waymo cars. (Id need to double check, but I think it's the LIDAR sensor that the tesla cars are missing). There will also be a lot of differences in the software of each car.
In reply to Rigid Raider:
Do you think most cars have the ability to follow the car in front by radar? Maybe, possibly, most brand new cars. But I doubt it's most cars on the roads. And until it is most cars (what is most?) it's difficult to see how smart roads will work.
Do you think most cars have the ability to follow the car in front by radar? Maybe, possibly, most brand new cars. But I doubt it's most cars on the roads. And until it is most cars (what is most?) it's difficult to see how smart roads will work.
1
In reply to remus:
Indeed. Tesla have strongly maintained that cameras and RADAR are sufficient.
It is obvious that in the long term they are correct (human drivers don’t have RADAR or LIDAR...) But LIDAR certainly makes the programming easier right now. Costs a fortune although miniaturised systems using phased array transmitters are in the works.
The interesting thing to me is the way Tesla are now using specular reflections of RADAD signals to measure the speed of cars more than one ahead. Many human drivers do this sort of thing all the time with light - awareness of an oncoming car round a bend or obscured traffic lights because of light scattering off a house, parked car, road surface or tree. This I think will be very hard for a passive system (cameras) to achieve but much easier for active ones (LIDAR and RADAR).
> (Id need to double check, but I think it's the LIDAR sensor that the tesla cars are missing).
Indeed. Tesla have strongly maintained that cameras and RADAR are sufficient.
It is obvious that in the long term they are correct (human drivers don’t have RADAR or LIDAR...) But LIDAR certainly makes the programming easier right now. Costs a fortune although miniaturised systems using phased array transmitters are in the works.
The interesting thing to me is the way Tesla are now using specular reflections of RADAD signals to measure the speed of cars more than one ahead. Many human drivers do this sort of thing all the time with light - awareness of an oncoming car round a bend or obscured traffic lights because of light scattering off a house, parked car, road surface or tree. This I think will be very hard for a passive system (cameras) to achieve but much easier for active ones (LIDAR and RADAR).
1
In reply to wintertree:
It's interesting the things we 'know' while driving without actually knowing exactly how we know them if we think about it. I think with experience we learn to subconsciously process a lot of very subtle cues which don't add up to enough for us to recognise or identify an actual thing but it is enough to put us on alert. On the other hand some of it is probably just confirmation bias, we slow where there may be a hazard we can't yet see then congratulate ourselves when there is one forgetting the hundreds of times there wasn't. It's not easy to unpick.
jk
> The interesting thing to me is the way Tesla are now using specular reflections of RADAD signals to measure the speed of cars more than one ahead. Many human drivers do this sort of thing all the time with light - awareness of an oncoming car round a bend or obscured traffic lights because of light scattering off a house, parked car, road surface or tree. This I think will be very hard for a passive system (cameras) to achieve but much easier for active ones (LIDAR and RADAR).
It's interesting the things we 'know' while driving without actually knowing exactly how we know them if we think about it. I think with experience we learn to subconsciously process a lot of very subtle cues which don't add up to enough for us to recognise or identify an actual thing but it is enough to put us on alert. On the other hand some of it is probably just confirmation bias, we slow where there may be a hazard we can't yet see then congratulate ourselves when there is one forgetting the hundreds of times there wasn't. It's not easy to unpick.
jk
2
In reply to JoshOvki:
"If everyone followed the highway code to the text, the only crashes would be mechanical failure"
and myself and at least two others on the mini-roundabout would still be deciding whose right of way it is...
"If everyone followed the highway code to the text, the only crashes would be mechanical failure"
and myself and at least two others on the mini-roundabout would still be deciding whose right of way it is...
4
In reply to LeeWood:
I admit I was a bit ahead of myself in claiming that "most" cars have the ability to follow the car ahead by radar but I believe it's the case with all VAG cars that are equipped with cruise control.
My experience of the technology is that it's pretty raw; my Passat reacts late to changes in speed of the car in front and lacks the anticipation that a human would have when s/he sees the cars ahead slowing, meaning that braking tends to be a little abrupt and progress is not as smooth as it can be with an experienced, anticipative human driver. I'm guessing that on a smart motorway the technology will take speed and spacing away from the car and the driver will only need to steer and monitor.
I admit I was a bit ahead of myself in claiming that "most" cars have the ability to follow the car ahead by radar but I believe it's the case with all VAG cars that are equipped with cruise control.
My experience of the technology is that it's pretty raw; my Passat reacts late to changes in speed of the car in front and lacks the anticipation that a human would have when s/he sees the cars ahead slowing, meaning that braking tends to be a little abrupt and progress is not as smooth as it can be with an experienced, anticipative human driver. I'm guessing that on a smart motorway the technology will take speed and spacing away from the car and the driver will only need to steer and monitor.
In reply to LeeWood:
BMW have released a model which aims to deliver close passes....
http://road.cc/content/news/230853-semi-autonomous-bmw-will-%E2%80%98fight-...
BMW have released a model which aims to deliver close passes....
http://road.cc/content/news/230853-semi-autonomous-bmw-will-%E2%80%98fight-...
In reply to LeeWood:
I thought autonomous cars had driven enough (millions) to show they're safer than humans, but...
... the driver takes over every 5000 miles (Google in 2016).
If whilst driving I only had to do something every 5000 miles I think I'd be asleep or not paying attention so I think I'd crash.
That doesn't seem to happen for Google though so maybe their drivers are more on the ball than I would be or the disengagement of autonomous control and takeover by a human is fairly benign process rather than an emergency.
In an emergency the Google car probably reacts earlier than an alert human anyway.
https://www.dmv.ca.gov/portal/dmv/detail/vr/autonomous/disengagement_report...
I thought autonomous cars had driven enough (millions) to show they're safer than humans, but...
... the driver takes over every 5000 miles (Google in 2016).
If whilst driving I only had to do something every 5000 miles I think I'd be asleep or not paying attention so I think I'd crash.
That doesn't seem to happen for Google though so maybe their drivers are more on the ball than I would be or the disengagement of autonomous control and takeover by a human is fairly benign process rather than an emergency.
In an emergency the Google car probably reacts earlier than an alert human anyway.
https://www.dmv.ca.gov/portal/dmv/detail/vr/autonomous/disengagement_report...
Post edited at 12:54
In reply to LeeWood:
Of course its not just a question for cyclists - children and animals too with all their undpredictability. Presumably they would be programmed to ignore a cat or crow (better safeguard human life) but not a ... donkey, cow, horse ?? At want point does size matter - obviously hitting a larger animal would nevertheless risk human life in the shunt.
Of course its not just a question for cyclists - children and animals too with all their undpredictability. Presumably they would be programmed to ignore a cat or crow (better safeguard human life) but not a ... donkey, cow, horse ?? At want point does size matter - obviously hitting a larger animal would nevertheless risk human life in the shunt.
1
In reply to elsewhere:
My volvo has a thing called Pilot Assist which steers as well as follows the car in front. You have to give a little bit of a wiggle to the steering wheel every ten seconds and it’s only really suitable for straight roads and shallow angled bends (motorways, dual carriageways). But it does make long drives more relaxing, though I’d agree that because it can’t see cars more than one ahead it doesn’t anticipate slowing traffic as well as a driver.
My volvo has a thing called Pilot Assist which steers as well as follows the car in front. You have to give a little bit of a wiggle to the steering wheel every ten seconds and it’s only really suitable for straight roads and shallow angled bends (motorways, dual carriageways). But it does make long drives more relaxing, though I’d agree that because it can’t see cars more than one ahead it doesn’t anticipate slowing traffic as well as a driver.
In reply to LeeWood:
Why presumably run over cats and birds? People value animals and their vehicles so simple commercial pressure suggests self-driving cars will take avoiding action to preserve life and bodywork where safe and possible to do so. Much like humans really but with he benefit of people having considered the rules for 'when safe to do so' in advance. Without fright and its disabling physical and mental effects they have the ability to process and react to dangerous situations promptly and objectively which many human drivers can't when it comes to (the moment before) the crunch.
I wonder whether any of the self driving systems gather data on surface quality, weather and available grip then make predictions for the road ahead which would feed into collision avoidance decisions? They potentially have the same cues available as a human driver: visual, temperature, control feel, past experience. Can they make sense and use of that data?
jk
> Of course its not just a question for cyclists - children and animals too with all their undpredictability. Presumably they would be programmed to ignore a cat or crow (better safeguard human life) but not a ... donkey, cow, horse ?? At want point does size matter - obviously hitting a larger animal would nevertheless risk human life in the shunt.
Why presumably run over cats and birds? People value animals and their vehicles so simple commercial pressure suggests self-driving cars will take avoiding action to preserve life and bodywork where safe and possible to do so. Much like humans really but with he benefit of people having considered the rules for 'when safe to do so' in advance. Without fright and its disabling physical and mental effects they have the ability to process and react to dangerous situations promptly and objectively which many human drivers can't when it comes to (the moment before) the crunch.
I wonder whether any of the self driving systems gather data on surface quality, weather and available grip then make predictions for the road ahead which would feed into collision avoidance decisions? They potentially have the same cues available as a human driver: visual, temperature, control feel, past experience. Can they make sense and use of that data?
jk
1
In reply to LeeWood:
Regardless of the technology, I'm not sure that autonomous (self-steering) vehicles are legal for use on UK roads.
Anyone know for certain either way?
Regardless of the technology, I'm not sure that autonomous (self-steering) vehicles are legal for use on UK roads.
Anyone know for certain either way?
In reply to paul_in_cumbria:
The twonk mowed down and killed the cyclist from behind.
> For completeness due to lack of evidence, that assumption could also apply to the cyclist couldn't it in all fairness?
The twonk mowed down and killed the cyclist from behind.
3
In reply to jkarran:
When the self drive car can factor into its intelligence the "two abreast cycling whatever the conditions" ingredient then I will be truly impressed.
When the self drive car can factor into its intelligence the "two abreast cycling whatever the conditions" ingredient then I will be truly impressed.
1
In reply to jkarran:
It all sounds v clever - I'm beginning to imagine a sensitivity / threshold control for roadkill ... select '0' to save sparrows
But seriously - I'm reassured to know its all so researched and trialled. Presumably the failsafe would force control back onto the driver in the event of malfunction. Whatever accidents might exist while under 'self-drive' one must judge this outcome against the accidents which continue to occur under direct human control. Just a shame that an 80yr old had to be victim the development history.
What is the additional cash outlay for such sophisticated control ? And who is going to choose that - or will it become mandatory ?
> Why presumably run over cats and birds?
It all sounds v clever - I'm beginning to imagine a sensitivity / threshold control for roadkill ... select '0' to save sparrows
But seriously - I'm reassured to know its all so researched and trialled. Presumably the failsafe would force control back onto the driver in the event of malfunction. Whatever accidents might exist while under 'self-drive' one must judge this outcome against the accidents which continue to occur under direct human control. Just a shame that an 80yr old had to be victim the development history.
What is the additional cash outlay for such sophisticated control ? And who is going to choose that - or will it become mandatory ?
In reply to captain paranoia:
With a driver on board, yes. http://www.bbc.co.uk/news/business-42024880
> Regardless of the technology, I'm not sure that autonomous (self-steering) vehicles are legal for use on UK roads.
With a driver on board, yes. http://www.bbc.co.uk/news/business-42024880
In reply to LeeWood:
I have a Tesla, it took a while to get used to taking my hands off the wheel on a motorway I would not trust it in a congested urban environment especially when there are a number of cyclists around as tests have shown that the current systems struggle to identify them and most cyclists in London rely on me to protect them and do as much as possible to test my alertness.
I have a Tesla, it took a while to get used to taking my hands off the wheel on a motorway I would not trust it in a congested urban environment especially when there are a number of cyclists around as tests have shown that the current systems struggle to identify them and most cyclists in London rely on me to protect them and do as much as possible to test my alertness.
Post edited at 10:39
2
In reply to Dogwatch:
I think that only applies for special licence for engineering development. I don't think it's carte blanche for anyone to let a car steer itself.
I think that only applies for special licence for engineering development. I don't think it's carte blanche for anyone to let a car steer itself.
In reply to Rigid Raider:
If he was involved in any such activity then this genre of car is still put to question in allowing the driver to imagine he can avert attention.
And whatever the outcome of the inquiry, a self-drive will surely open an unprecedented can of worms in the legal-machine - who/what exactly was at fault in this situation ??
> The article says that it's not known whether the car was in self-driving mode or not at the time of the accident. The driver could have been texting, asleep, drunk, half blind or distracted.
If he was involved in any such activity then this genre of car is still put to question in allowing the driver to imagine he can avert attention.
And whatever the outcome of the inquiry, a self-drive will surely open an unprecedented can of worms in the legal-machine - who/what exactly was at fault in this situation ??
In reply to captain paranoia:
I can't see that manufacturers would be selling cars that can to an extent self-drive if it weren't legal - as I said upthread mine steers itself but every ten seconds you need to let it know you're still there.
> I think that only applies for special licence for engineering development. I don't think it's carte blanche for anyone to let a car steer itself.
I can't see that manufacturers would be selling cars that can to an extent self-drive if it weren't legal - as I said upthread mine steers itself but every ten seconds you need to let it know you're still there.
In reply to LeeWood:
One of the designers has said that the self drive should not be employed out of motorway situations. I would agree with that, keep it to the places where it is just cars without vulnerable road users. The drivers need to be fully alert and paying attention when vulnerable road users are present and during self driving that would not be the case.
One of the designers has said that the self drive should not be employed out of motorway situations. I would agree with that, keep it to the places where it is just cars without vulnerable road users. The drivers need to be fully alert and paying attention when vulnerable road users are present and during self driving that would not be the case.
1
In reply to Rigid Raider:
I have a VAG car with cruise control and I assure you that it doesn't have that capability.
I think we're some way from having a high enough proportion of platooning-capable vehicles on the the road to make the technology useful or reliable.
> I admit I was a bit ahead of myself in claiming that "most" cars have the ability to follow the car ahead by radar but I believe it's the case with all VAG cars that are equipped with cruise control.
I have a VAG car with cruise control and I assure you that it doesn't have that capability.
I think we're some way from having a high enough proportion of platooning-capable vehicles on the the road to make the technology useful or reliable.
In reply to Dark-Cloud:
And linked from that very article is one which reports findings that the Tesla system can't see cyclists:
http://road.cc/content/news/223386-never-use-tesla-autopilot-feature-around...
> BMW have released a model which aims to deliver close passes....
> http: //road.cc/content/news/230853-semi-autonomous-bmw-will-‘fight-driver’-deliver-close-passes-cyclists
And linked from that very article is one which reports findings that the Tesla system can't see cyclists:
http://road.cc/content/news/223386-never-use-tesla-autopilot-feature-around...
2
In reply to kmhphoto:
From the article in the OP: " Tesla says it encourages Model S drivers to stay alert and keep their hands near the steering wheel"
You may as well drive the damn thing yourself!
And if you do need to react, there's an instant delay, 1: decide to react, 2: get your hands on the wheel.
> I have a Tesla, it took a while to get used to taking my hands off the wheel on a motorway I would not trust it in a congested urban environment
From the article in the OP: " Tesla says it encourages Model S drivers to stay alert and keep their hands near the steering wheel"
You may as well drive the damn thing yourself!
And if you do need to react, there's an instant delay, 1: decide to react, 2: get your hands on the wheel.
2
In reply to Martin W:
supports my vote for a green flashing beacon !!
> And linked from that very article is one which reports findings that the Tesla system can't see cyclists:
supports my vote for a green flashing beacon !!
In reply to andy:
Tesla is a very strange situation. They sold plenty of cars with self-steering capability before it was legal to rely on it (in the US). They told buyers not to take their hands off the wheel. Of course, drivers did, and posted videos of themselves doing so on YouTube. Including videos where disaster was only narrowly avoided when they took control after the system went haywire.
Tesla released software they called 'beta test'. You DON'T release beta safety critical software. Essentially, they are using their buyers as engineering testers.
I'm an engineer. I'm not opposed to the idea in principle. I just don't think it's mature enough yet.
Tesla is a very strange situation. They sold plenty of cars with self-steering capability before it was legal to rely on it (in the US). They told buyers not to take their hands off the wheel. Of course, drivers did, and posted videos of themselves doing so on YouTube. Including videos where disaster was only narrowly avoided when they took control after the system went haywire.
Tesla released software they called 'beta test'. You DON'T release beta safety critical software. Essentially, they are using their buyers as engineering testers.
I'm an engineer. I'm not opposed to the idea in principle. I just don't think it's mature enough yet.
2
In reply to captain paranoia:
The technology is definitely not mature enough but its used as a marketing tool to entice buyers and investors. I have an "S" model and would now say its at least £50K over priced so the marketing is very effective
The technology is definitely not mature enough but its used as a marketing tool to entice buyers and investors. I have an "S" model and would now say its at least £50K over priced so the marketing is very effective
1
1
In reply to kmhphoto:
Tempted to sell it on for (£50k under retail price) depreciated at -20% per year for its age? Drop me a line...
> I have an "S" model and would now say its at least £50K over priced so the marketing is very effective
Tempted to sell it on for (£50k under retail price) depreciated at -20% per year for its age? Drop me a line...
2
In reply to LeeWood:
I took delivery of a Tesla P100D recently, and all I can say is that it is a stunningly sophisticated car, and has made decisions and adjustments in a nano second, and before I've even spotted what it was making adjustments and decisions for?
I can only base my comments on a relatively short time with it, but I have little doubt that it is certainly ten times faster, smarter and safer than my reflexes and decision making abilities.
It is also possibly, the most remarkable car I've ever driven.
I took delivery of a Tesla P100D recently, and all I can say is that it is a stunningly sophisticated car, and has made decisions and adjustments in a nano second, and before I've even spotted what it was making adjustments and decisions for?
I can only base my comments on a relatively short time with it, but I have little doubt that it is certainly ten times faster, smarter and safer than my reflexes and decision making abilities.
It is also possibly, the most remarkable car I've ever driven.
2
In reply to LeeWood:
I have a VAG car and it's ACC is great but detecting cyclists is a little hit and miss (no pun intended). Motorbikes and mopeds are generally detected, however I'd be rather nervous passing horses with it engaged.
I have a VAG car and it's ACC is great but detecting cyclists is a little hit and miss (no pun intended). Motorbikes and mopeds are generally detected, however I'd be rather nervous passing horses with it engaged.
1
This topic has been archived, and won't accept reply postings.
Loading Notifications...