
The bad news for Musk and his minions just keeps on a' coming. That's what happens when it is eventually exposed that you're nothing more than an egotistical and mouthy flim-flam man who doesn't pay his bills...
A Tesla driver was fast asleep at the wheel with Autopilot engaged during a 15-minute pursuit by police, the cops claim. Traffic officers said they spotted the vehicle as it traveled east on the autobahn from Bamberg to Bayreuth, Germany. The patrol signaled to the car to stop but according to Bavarian police, the 45-year-old …
I don't know anything about space rockets, so to listen to Elon Musk speak about them, I think he's a genius.
I don't know anything about electric cars, so to listen to Elon Musk speak about them, I think he's a genius.
I do know at least something about tech stacks, website coding and web hosting servers, so when I listen to Elon Musk speak about them, I realise he's a freaking charlatan.
As a result, I now have zero faith in his space rockets or electric cars.
At least Stephen Fry has a decent sense of humour and can handle satire..
I followed the development of SpaceX's Falcon 9 and it's landing capabilities with much anticipation over the last few years. I quickly realised that it was worth paying attention to articles / media showing actual progress (grasshopper trials, test flights etc), but there was very little point in paying attention to what Musk was saying. More often than not, most of what he announced was vapourware - an end goal that might or might not be reached.
..."I followed the development of SpaceX's Falcon 9 and it's landing capabilities with much anticipation over the last few years. I quickly realised that it was worth paying attention..."
Especially when the first series of first stage rockets attempted to land...and I knew that such a tall, slender rocket could never land perfectly and would fall over - and they did, usually leading to a "big booooom" and lots of small bits of rocket spread over a wide area !!
I always thought they should have had some form of "outrigger" system that deployed, that kept the rocket vertical when it landed - and thankfully after all the first attempts to land failed, someone at SpaceX also had the same idea of an "outrigger".
"Those were tests. They weren't meant to survive."
Yes they were tests...but most rockets can go up fairly reliably as the technology has been know since Goddard in the 1920 and the Chinese since 2,000+ years ago.
The trick that SpaceX tried to pull was landing a rocket...and "those tests" covered various factors such as "guidance" (to accurately land on the big "X" marked on the ground) and for the rocket to be reusable after landing.
The cheers from the watching crowds at SpaceX as the rocket was landing was huge...but it was quickly followed by groans as the thing fell over and exploded. This happened many times. So you would think they would quickly design something to stop them falling over - and eventually they did.
This post has been deleted by its author
Musk, and even SpaceX, are not in the business of pleasing spectators.
It is however Musk's to manage Investors and other 'stakeholders'. What we, and even many scientists skilled 'in the space' think, matters relatively nought.
Your comments appear to be about PR and public opinion about one of the testing programs in which he only seems to be involved- but actually involves spin doctors.
Plenty of madness in the space to poke fun at of course. Criticism of Musk relating to his job would probably be received a little better however, even here on El Reg.
Someone wasn't watching the preceding grasshopper program very attentively then.
Before they ever attempted re-entry burns or soft landings, they had a couple of test vehicles with landing leg gantries attached. These vehicles started with short hops of a few metres, literally enough just to take the weight off the legs, then worked up to hops of about a kilometer.
A wikipedia link, for your perusal: https://en.wikipedia.org/wiki/Falcon_9_prototypes
As a result, I now have zero faith in his space rockets or electric cars.
You can have faith in his space rockets, they're well proven now. (SpaceX has launched more orbital rockets than everyone else combined, they managed about 2 a month last year).
Gwynne Shotwell doesn't really get the credit she deserves.
Elon is like all these visionaries, forceful enough of personality to make things happen. It's what you need for things like SpaceX and Tesla to get going. (But not for running more run of the mill companies like twitter, that are large and significant but not technially that groundbreaking).
Also, Tesla do like to over hype their marketing of the self driving stuff, but for fucks sake if you're activly using a device to defeat the cars safety systems then surely that should be the same as intentionally driving a vehicle you know isn't roadworthy.
I don't know anything about space rockets, so to listen to Elon Musk speak about them, I think he's a genius....
This is essentially what Michael Crichton called the "Murray Gell-Mann Amnesia Effect": the tendency to be critical of a source only when we have enough expertise to judge that source with some accuracy, and then revert to giving it credibility when it's discussing other topics. It's an under-appreciated member of the large family of human cognitive traps.
"The bad news for Musk ..."
What, that someone was deliberately circumventing the control system of the vehicle? No more Musk's fault than Fords being "unsafe" if someone put a brick "cruise control" on the throttle of a Ford Focus or a if a 17year old drove like a pillock in a chipped Renault Clio ... whatever one of those is.
I'd say it's more a question of roadworthiness. A Ford Focus doesn't have a built-in "self drive" feature that relies on having your foot on the accelerator, and a car manufacturer is not responsible for bad or illegal driving.
Teslas, however, have a "feature" that is inherently dangerous is misused, and market this in a deceptive manner (calling it full self-drive or whatever when it is just fancy cruise and lane control). If there's a risk it can be accidentally activated, or that measures to ensure that it is not misused are easily countered, then there is a question of whether this "feature" actually makes the vehicle dangerously unroadworthy.
So a car manufacturere is not responsible for bad or illegal driving, except when they are??
or that measures to ensure that it is not misused are easily countered
Tesla needs to take measures to ensure it's not accidentally activated, but once a driver conciously tries to circumvent the safety systems then the onus is on them. This is true if the car is a Tesla or a Ford or whatever.
If you get into a car and turn the traction control off then stuff it as a result, that's not the cars fault.
Tesla do need to stop marketing this stuff as full self driving, but in this instance it doesn't sound like they're at fault.
It is quite possible, of course, to modify a vehicle to make it unroadworthy (this is not the manufacturer's fault), such as removing the airbags, or welding spikes onto the bonnet.
Features on a vehicle, however, should "fail safe". If I'm driving my car down the motorway at "70" mph, and attach a weight to the steering wheel, and take my feet off the accelerator, it will fail safe by coming to a stop; albeit being stopped on a motorway isn't exactly safe, but it's safer than continuing to travel at 70 mph until it hits something.
The "auto drive" feature on a Tesla (or on any other car) is not failing safe in this case, and the effort to make it "fail dangerous", and effectively continue to travel under no control is apparently trivial. If the failure modes can't be made to be overwhelmingly safe, then the feature is dangerous.
No more Musk's fault than Fords being "unsafe" if someone put a brick "cruise control" on the throttle of a Ford Focus
Ah, well, you see, the Ford Focus has a "Will Crash" feature that guarantees you will eventually have a bad day if you don't pay attention when driving. It's 100% reliable.
The Tesla is exactly the same, except Telsa called "Autopilot" and it might take a little bit longer for you to crash. But you are definitely 100% guaranteed crash eventually, if you persist with not paying attention. You just have to be more patient, compared to the near instant gratification dished out by the Focus.
Teslas on autopilot are programmed to ignore stationary objects - so they run into parked emergency vehicles or barriers between the carriageway and a slip road if the lane marking are worn.
If a police car were to get in front and then slow down then it might work - OR the Tesla might just pull out to overtake.
"If a police car were to get in front and then slow down then it might work - OR the Tesla might just pull out to overtake."
There's a TikTok going round of exactly this happening. 2 police cars try to box a Tesla in; the one in front starts to slow down and the Tesla tries once to overtake but when the manoeuvre is matched by the police car, it gives up and falls in behind it, then slows to a standstill.
No idea of the circumstances, the comments were full of the usual crap but seemed to be a case where the driver was taken ill rather than any serious attempt to escape.
You're assuming way too much sophistication, it's a Tesla.
Apart from FSD they're about as basic as you can get* - and FSD doesn't even work properly.
Heck, it can't even park properly.
* As a result of the same desperate attempts at cost saving that also made Musk forgo anything but cameras for environmental detection, the primary reason Tesla's FSD misses so much.
Interesting that the Tesla didn't even to try to park in a UK standard sized car park space and made a dogs dinner of the parallel parking. If I'd bought a Tesla and the extra for self-parking, I'd be very disappointed.
I suppose it's not been properly "trained" on UK parking and is assuming larger US parking spaces.
Indeed, the most direct way to stop this vehicle would have been parking an emergency vehicle with all its lights flashing a bit further down the road because the Tesla would have immediately behaved like an Ukrainian missile seeing a cluster of Russian military and crashed into it, but apart from damaging the Tesla (which, granted, would have less of an issue other than the inevitable unquenchable battery fire), it would also have damaged an emergency vehicle.
I wonder what would have happened if they simply got in front of it without the Tesla-confusing emergency lights and started slowing down. It would leave enough time to get out of the way if the vehicle didn't react but normally that part is at least supposed to work even on Fools Self Driving..
This post has been deleted by its author
"Assuming Autopilot was engaged, wouldn't the Tesla have come to a standstill if the Police had got in front of it and then decelerated to a standstill?"
I wouldn't try it. It might work, but the car would probably try to pass them if it could. If anything goes wrong, the car crashes into the police, potentially causing serious injury to the driver and maybe even the police, and certainly wrecking two cars. Given the tradeoffs, I probably wouldn't take for granted that the system wouldn't make a mistake. Unless there was already another risk to safety, I would use the police's methods and follow it to watch and interfere when possible.
This post has been deleted by its author
This post has been deleted by its author
This post has been deleted by its author
This post has been deleted by its author
This post has been deleted by its author
Train drivers can often be seen carrying a large bag with a shoulder strap. Inside are a couple of weighty tomes that detail for each possible type of accident who is to blame.
The dead man lever on some trains has such a strong spring that it is impractical to keep it held down for anything like a full shift.
It is possible these statements are in some way related.
The railway companies realised the risk of weighing down the 'Dead Man's Pedal/Handle' years ago and modern trains instead have a pedal that has to be released and depressed again in response to a vigilance alarm (you can often hear the 'Beep beep beep thunk' if you're sitting behind the driver's cab).
I have heard mention of an incident in the past where apparently an Underground train ran off without the driver. Apparently he left his bag weighing down the handle and went back to fix an issue with a door that wasn't shutting properly. As soon as he fixed it and the door closed properly, the train gained traction interlock and set off without him.
It has to be far easier to monitor the driver than all the other road users so why isn't this a legal requirement?
If the "AI" can spot a cyclist or someone lying in the road then why can't it tell that the driver is sitting in the correct position with their arms extended and their hands on the wheel. Eye tracking isn't exactly new tech why isn't it being used to tell that the driver is awake and paying attention? If you say it's coz of "shades man" then I'm sure they could sell a range of suitably logo'd Tesla shades with built in sensors and eye tracking.
> You're assuming that the "AI" can spot cyclists and people lying in the road.
Nope!
No I'm not, but a functional driving system needs to be able to and I'm just saying that monitoring that the driver is behaving is easier than monitoring the rest of the real world.
But
Monitoring the drivers behaviour probably doesn't rate highly as a sales tool and so unless forced by law to do this then there is no impetus for Tesla or any other manufacturer to fit a feature like this.
This post has been deleted by its author
GM, etc, along with an all-cabin camera, have a camera mounted in the dash pointing straight at the driver, and it can handle Near Infra-Red, so it can see through sunglasses and do all that fancy eye-tracking stuff. However, Tesla only has one non-IR camera mounted below the rear-view mirror that covers all the cabin, so at best they can only see one eyeball, I would guess. Even so, it should be more than possible to detect the driver lying back, or whatever, but as we see elsewhere, features seem more important that safety in Tesla.
This post has been deleted by its author
Good morning. The car's interior temperature is 16 degrees, the estimated driving time is 34.7 minutes. Mrs Müller, you weigh 28 kilos more today than last night. Mr Müller, should I put the passenger seat in the reclining position, like last night?
An EU directive (yay Brexit???) coming into force soon says that all driver monitoring must be processed locally and deleted as soon as handled, but I'm not sure how it handles that it still has to send out a signal that can be logged such as DRIVER_SLEEPING, DRIVER_INATTENTIVE, DRIVER_WATCHING_PR0NHUB...
Don't see an issue - all driver monitoring is processed locally, sending out a 'status' signal doesn't change that - provided the drive monitoring process doesn't listen for a remote response to the 'status' signal and take action based on any such response.
The key word is 'processing' not 'data'...
.. which is exactly the potential FSD has and why it should not even be on the public road.
And no, the excuse that it hands back to the driver when it gets confused doesn't work - (1) if the driver was paying attention, why would he/she even need FSD and (2) quickly handing off when it gets hairy may avoid liability, it doesn't improve safety, IMHO it makes things worse by providing a false feeling of safety. Oh, and (3) never make something available for a public road that idiots can abuse, because they will, without fail.
.... is that if the driver is supposed to be alert and with their hands on the wheel .... why can't they just drive it instead?!?
This pretext of the driver must be holding the steering and alert while the 'autopilot' is engaged is just a load of bovine excrement! Either the car can drive itself or it can't. As soon as you give the average numpty one of these things they'll be careful for the first few trips, but then once they've seen how well it drives (in normal circumstances) they'll trust it more and more to do the job for them. The first time that something 'unusual' happens the autopilot is gonna fail to deal with it and it'll all depend on how much (or little) attention the driver is actually paying.
So lets call this what it actually is .... a bait and switch. And one that's gonna kill more people before some little kid stands up and says the "Emperor has no clothes on"?
is that if the driver is supposed to be alert and with their hands on the wheel .... why can't they just drive it instead?!?
Have you tried it? In some circumstances it really does help with long drives, the precise lane keeping and steady distance keeping is lifted from you leaving you to just watch for hazards and monitor the other traffic.
I just find it unusable when the road is busy because the actions of other drivers are constantly affecting it.
Audi's, BMW's, Mercs - there's a thing with prestige or 'sports' vehicles that attract a particular type of driver. Doesn't mean all drivers in said vehicles are bad, just that they attract bad drivers.
Had my own incident (M6) with a BMW that didn't understand the motorway ahead was stationary, as it usually was at that time of day, on a Friday, and to slow, even stop. 70mph into the boot of a Volvo - which really was a solid lump of metal and took the impact better than the BMW.
However, bad drivers will be bad no matter what car they're driving.
I had my incident over a decade ago, when the driver of the Audi rammed into the stationary Volvo that was behind me on a moped, waiting to turn right on a narrow country road.
The front of the Audi was completely destroyed, the moped was written off by the insurers (I suspect the personal injury claim cost the Audi driver more money), and the Volvo was pretty much unscathed.
I overheard the Audi driver trying to claim to the driver of the Volvo that they shouldn't have been stationary in the road, and bemoaning her fate that she was on the way to sell the car she was driving (my heart bleeds). Just the fact that she was travelling faster than she was able to stop in the distance she could see*, and thought that this was okay, is deeply worrying. People like that should not be on the roads.
*Tcehnically, it should be half that distance, because you might wish to avoid a collision with someone coming the other way.
(Posted anon, as this story provides far too much identifying information about me, and I'm supposed to be hard at work)
"70mph into the boot of a Volvo - which really was a solid lump of metal and took the impact better than the BMW."
To be fair, crumple zones are designed in at the front, less so at the back. It being a Volvo that was rear-ended just adds to the lack of crumple at the back :-)
>Have you tried it?
Yes, frequently.
However, this is probably part of the problem, the average driver in the UK only does circa 6000 mile pa - mostly as short around town journeys, so I've built up the concentration stamina etc.
>In some circumstances it really does help with long drives
For normal people, I suggest they would be better of taking a break.
I seem to remember that you (werdsmith) have a physical disability and hence whilst you may have the mental stamina you have good reason to find long drives a bit more of a physical challenge.
Have you tried it?
Yep. In many a busy city rush-hour. Frankly, if there was a risk the vehicle might crap out over some coned-off roadworks or defaced road markings or stationary broken-down vehicle in the not-the-hard-shouder-right-now lane, I would disable the bitch and take over for the duration.
In some circumstances it really does help with long drives.
A safety backup that relaxes its attention so it doesn't get tired is not being a safety backup right then. Sheesh!
The idea is that eventually the cars will drive themselves as people aren't to be trusted driving, or with anything else.
That the human brain is still far superior to Artificial Idiocy, and human senses are more adaptive and capable than those fitted to cars (often of the lowest quality to save money!) they feel that the devices should be fitted and used, but the human driver will be responsible and take the blame for any 'accident' the system causes. This is what we get with the break assist and other 'safety features' in cars in general: They're systems that build up into an autonomous vehicle, but they're not ready for that yet: Doesn't stop governments mandating they're fitted to all new cars to push the tech forwards - and it doesn't stop incidents when the tech gets things wrong and causes an incident (and then blaming the driver, because there has to be a driver in charge of the vehicle, even if the vehicle overrides the driver's control).
This doesn't just apply to Tesla, but Tesla are pushing this tech and trying to run well before they can even crawl, and this is why, when people misuse this equipment (which is good enough for a steady drive on a straight road with no hazards - just) it's still the human who is to blame (according to Tesla and anyone else pushing this tech).
Autopilots have only ever been 'good enough' for routine, mundane travel where humans get bored and can get distracted. They fail spectacularly, and often fatally, where there's something to adapt to, such as a mountain in their path, or turbulence, or a pothole. Or a lorry pulled across the road that the front sensor can't see...
I'll give Musk and Tesla (and his other enterprises) credit, though: They're willing to give things a try. That they're not always successful (or sensible, or even sane) just adds to the entertainment, as long as people don't get hurt in the process.
Personally, I'd rather the car work with the driver, not against them. I'd rather these systems advise or inform, not try to take control. Autopilot on a ship, or on a plane, work because they're supervised, and there aren't many things to run into (and systems should warn of anything ahead), but even then they mess up and aircraft hit mountains, and ships run ashore. I can see networking cars so they follow the leader, but the leader needs a human to be in control. But beyond that? Tech's no where near ready.
"lane-keeping is almost pointless because you have to keep your hands on the wheel, and naturally steer anyway."
Any Tesla drivers care to tell us how much pressure the steering wheel responds to? I sometimes drive with a very, very light touch when on a long straight nearly empty motorway, eg elbows on knees and hands barely touching the wheel at 20 past 8 position.
Let me add some clarifications to comments and questions elsewhere in this thread from people who, fair enough, are not tesla drivers and therefore couldn't know.
"Autopilot" is not a misnomer; like in an aircraft, the tesla autopilot is a driver assist feature. Just like an aircraft on autopilot won't land itself and won't automatically watch the airspace to replan its routes around busy airspace or towers, a tesla on autopilot will do a select few things that simply assist the driver. An aircraft autopilot will hold altitude, bearing and speed. A tesla autopilot will hold lane and a set max speed unless there is traffic in front, in which case it holds a safe distance. As a daily tesla driver who has also done cross-europe roadtrips I can say with authority that this driver assistance feature is indeed nice to have, it is an assistance feature that works as intended. Tesla never sold autopilot as any more than this - however, media has often made wrongful assumptions about what autopilot was meant to be. Tesla never sold autopilot as any more than a driver assist feature - which is what it is and does well.
Just like your aircraft pilot shouldn't do drugs or sleep when the plane is on autopilot, a tesla driver also needs to have hands on the steering wheel and to be awake and pay attention. This is very clear from the documentation of the system and the way it's implemented. Yes it's trivial to defeat if you want to - but you can drive a diesel Hummer while snorting coke too so what's your point?
If the emergency vehicles would have taken up the lanes in front of the tesla and simply slowed down, the tesla would have safely come to a full stop. End of story. Yes it is this simple. I drive in rush hour traffic every day in my own tesla, on autopilot, and I can state with authority that this is indeed how the system works. Pointing to articles about "evidence" saying otherwise which are really articles about stationary objects in the shoulder of the road are not on point here. This trivially could have slowed this car to a halt, had they known how. And as other carmakers are slowly catching up on driver assist features, this will be true for more and more makes of vehicles. Yes tesla is ahead, but all new vehicles today are getting driver assist features - this is not a tesla only feature long term.
I'm not claiming all tesla does is perfect. But they are pretty good - better than anything else I've seen and tried - and the publicly available statistics back this up too:
NHTSA's accident statistics for Q4 2021:
Teslas with autopilot recorded one crash for every 4.31 million miles
Teslas without autopilot recorded one crash for every 1.59 million miles
Average across all makes is one crash for every 484,000 miles
So basically switch from the 'average' non tesla to a tesla on autopilot and you decimate your chance of being in an accident. These are publicly available numbers.
Fair enough, that's cool but my point is that it is still a system that solves a limited set of problems. Like the tesla autopilot, your aircraft autopilot is not fully autonomous to the point that the pilot does not need to be present and lucid. Which was the point I was making.
An aircraft on autopilot will follow the flight plan and land itself if it's in approach mode.
You are getting confused, most autopilots on GA aircraft are height and heading hold systems, or just heading hold. I use one, I tell it what heading I want, and I have to calculate the wind offset to obtain the track I want. There is no FMS for me to input a flight plan and no chance of it flying an ILS. But it is most definitely an autopilot, manufactured by Bendix/King.
If you want to fly an automatic approach and landing then you need a cat 3 autoland system, not an autopilot, and you need to be landing on a cat 3 equipped runway. This is done for low visibility situations and in most cases the first officer will hand fly the flare and landing.
First of all, Tesla doesn't count an accident as autopilot related if it has handed off to the driver, and we've already flagged that that creates probably the most dangerous situation of all: a driver scrambling to regain situational awareness exactly at a time when the car has already steered him into danger. But, conveniently, Tesla gets itself the Shaggy defence as a consequence. Don't you just love US marketing and lawyers?
Secondly, this may help correct some of your stats, and that's just deaths alone.
thats some mighty revisionist crap
He called it fucking autopilot knowing full well what normal people would believe it would do.
fuck off with that retelling of history.
Musk used it for his hype bollocks.
Tesla fans are just the worst of humanity totally fooled by a fuckwit
It's not even Autopilot. That's actually named accurately, aircraft autopilot doesn't let the pilot just fuck off, it still requires full attention to flying, just like Tesla Autopilot.
What's deceptive is Full Self Driving. That implies you can just take a nap and let the car do the driving. And it's not that at all, it should have been illegal for him to sell that until it actually could.
Also a Tesla driver here (basic Autopilot, not FSD). It's very nice to drive but of course not perfect (what is?) and that doesn't make me a Musk-fan any more than a Ford driver would be a personal fan of the head of the company - so please don't generalise, people.
I have to say, aircraft Autopilot is not marketed at the general consumer, whereas Tesla is. There is a general perception of what "autopilot" means, which may be wrong but Tesla need to take that into account because it does mislead people. However "Full Self-Driving" is just out-and-out wrong. Clearly optimism got the better of them when naming these features.
I keep reading "FSD" and my brain keeps immediately jumping to Elite: Dangerous and "Frame Shift Drive".
But that then got me thinking, that game does have an "autopilot" feature for docking, undocking, takeoffs and landings, and the important thing to note is that, even in a game where the thing controlling the "autopilot" has full control over the environment as well, you'd be a fool to use it unattended, because it's not perfect, and sometimes requires you to take manual control to avoid collisions, or "lingering too long" in the docking port of a space station.
The thing, here, is that, in the real world, pretty much none of the variables and parameters are known, and nothing else in the universe is under the control of the "autopilot" system, turning what is probably (in a single player mode) an NP-hard problem into an impossibility. The idea that a "self drive" system can act predictably and consistently safely within an unbounded system sounds implausible at best. Humans don't always manage it, and our hardware is a damn sight more complex and adaptable than whatever his Muskiness can build into a glorified milk-float.
So basically switch from the 'average' non tesla to a tesla on autopilot and you decimate your chance of being in an accident. These are publicly available numbers.
No. Bad stats.
Teslas with autopilot are more often used on freeways & other 'wide' roads - fairly controlled environments (I do not think they are legal in the UK)
Manually driven cars are driven in city centres - an uncontrolled environment.
There is a much greater chance per mile of an accident in an uncontrolled environment.
Apples / Oranges.
You need to compare tesla without autopilot with other car that has a similar safety braking feature etc.
I think you will probably find they are about the same.
Look at GM - their SuperCruise system uses a camera to look for eyes looking ahead, it doesn't rely on some simple torque applied to the steering wheel (in fact, it's designed not to require your hands on the wheel at all).
This stuff needs to be standardised and made part of the regulations, to ensure the driver is paying full attention when required.
I don't necessarily disagree with you, but...
Most cars have 'cruise control' which will allow you to set a fixed speed which it then maintains. If you fall asleep the cruise control will drive your car into someone who happens to be in the way. This is decades old technology and as such a decades old problem.
More advanced systems that can slow down to maintain safe distance and maybe even bring your car to a stop are an improvement, since they in some cases will prevent you falling asleep at the wheel from causing someone to die. But we still have a problem of course.
Now, you're not supposed to drive if you're not lucid and able, driver assist features or not.
You could argue that *all* vehicles should have an advanced surveillance system like these select GM models to prevent people from not paying attention at the wheel. I see that you can make that argument. However, in fairness, this is not tied to driver assist features - in fact, it seems that this system would be even more warranted on vehicles that are *not* able to slow themselves down safely when you fall asleep (or do drugs or use your phone or ...).
That would mean mounting this on all vehicles, especially those without advanced driver assist features. But that would mean adding cost to the lowest cost vehicles. That means going after the little guy - those who have the least money to spend on transportation will be hit the hardest by such regulation. Which I suppose is why the popular argument is that only the advanced (expensive / premium) vehicles who have the least need for this should have a requirement for it. Ah, politics :)
Tesla are not the only ones working on automated driving. The others saw the problem and limited the technology they sold so that humans still have something to do that can keep them alert. Behind the scenes they are working on systems that do not require human supervision. For some reason, Tesla think is is fun to sell temporarily license software with a misleading name and sort of promise that it will be working real soon now.
Audi already had a self driving facility (but, granted, not on an occupied road) in 2010 when they sent a car up Pike's Peak without anyone actually steering and got a license for road testing in the same year that Tesla started to just talk about Autopilot, and I am very certain other car companies are working on this too.
However, precisely because they ARE car companies they will not throw this on the road without a lot of testing as they have a whole industry to lose, unlike Tesla.
You can see this because elements of self driving start to pop up in various places. Lane assist, break assist, cross road scanners, decently working auto parking in both parallel and perpendicular mode, even independent self parking into garages (and note that that one was already demonstrated 9 years ago). They just do a lot more testing.
I agree that we shouldn't have feature bloat across all cars, but there's an issue of expectations.
When someone gets in a car with basic cruise control, they have no expectation other than one of it holding speed.
Even with adaptive cruise control and lane assist, there's only an expectation of it slowing down, speeding up and holding you in a lane (to varying degrees of ping-pongingness). Even with that we're in ropey territory, and most people will have experienced failures of those systems often enough to not trust them implicitly.
However, when a car advertises magic (with the power of advertising), even though it's mostly a combination of the aforementioned features, people expect magic, and their brains seem to switch off. And it's at that point where there need to be regulations in place to stop people having too much faith in the technology.
Essentially where expectations far outstrip reality, we need some way of ensuring the consequences aren't dangerous. Marketing has run amok..
And that's a decent argument for including such tech in a vehicle to be deployed automatically in the event that the vehicle detects the driver is unable to continue driving the vehicle, with the proviso that when it deploys it'll do so solely for the purposes of getting the vehicle parked up safely ASAP whilst contacting the emergency services, and not so that it'll just plod on along the planned route as if nothing had happened unless forced to stop. It's rather less of an argument for having that tech available at all times for the user to choose to enable just so they can stop paying attention to the road ahead of them.
The issue here isn't so much the availability of tech like this, it's the way in which it's being implemented/promoted, and whilst other manufacturers have stuff which is somewhat similar, it only seems to be Tesla that have pushed the promotional boundaries to/beyond the limits and given rise to this subset of drivers who really do seem to think their cars will do it all for them.
A key consideration should be - and then what happens? If the Johnny-cab in question decides the lummox at the front is no longer capable (the weight fell off, the snore-detector went off, there is a blue flashing light in the e-mirror or something) and it is doing precisely 70mph (of course) in lane 3 then what next? Does it stop in the middle of the road? Cut across traffic and park? Just turn the autopilot off? etc. etc. It would seem some consideration of the correct actions that should be taken needs some thought, and to be documented in a standard that all manufacturers must adhere to. As more of these infernal contraptions are let loose it would seem helpful that they are at least predictable (and as a bonus do something sensible)!
The Tesla market and fan base in the USA is fundamentally weak, and a lot weaker than they would like people to know. It’s perfectly possible to set up the Autopilot system in their cars to disconnect without driver input or attention (just drive any Tesla in Europe and you’ll be nagged to make a steering input every 15 seconds or so), yet in the USA the system will allow frankly appalling risks to be taken. They could implement the European restrictions, but the increased safety will put off their target buyer.
If they really wanted. Just have a simple camera that detects your pupils and where they are looking, which thanks to modern smartphone economies of scale would add only a few dollars to the car's cost of production. If you go more than few seconds without having your eyes open and looking at the road, it sounds an alarm, that increases in volume the longer you keep your eyes closed or looking elsewhere.
But Musk doesn't want to implement something like that, because he knows his customers are misusing Autopilot/FSD and knows that is part of the appeal of the car for some. So they make token changes trying to detect hands on the wheel, or detect cheating devices but they are making these changes with a wink at the customers using them leaving gaping holes for them to continue to defeat them.
Am I missing something? Why would a police car not just get in front of the Tesla and slow down? Surely the Tesla would see the slow moving traffic in front and stop (I really hope it would!), or did it mount the pavement and drive through the tables and chairs on the sidewalk like they do in so many American films?
I'm guessing California will now demand an SMS gateway so they can text Tesla a vehicle registration number and ZIP code, and have the vehicle self drive to the police pound :-)
This post has been deleted by its author