This almost got me into a serious accident, twice. Definitely would not recommend and I switched cars because of it. The last thing I need is buggy software with the capability of instantly changing the trajectory of my vehicle (slippery surfaces, cornering). A brand car as well (MB) so really, really disappointed, what should have been a safety feature actually was the opposite. Dealer said everything was fine so that was the end of that car, if 'fine' means that you have a couple of close calls in the first few thousand km on account of a feature that can't even be properly disabled without sabotaging the electronics then that's just too bad.
Now I drive a 27 year old car that has none of these 'features'. It does what I tell it to and when I tell it to, it doesn't phone home, and has no touch screen, but old fashioned buttons.
If the government wants to immediately improve car safety they should outlaw touchscreens instead.
I wrote up a long comment that got scrapped by my focus filter but I feel the need to voice strong agreement here. I have a recent Mazda which in almost all ways is my perfect car except for the newfangled safety features all new cars seem to have. Lane correction and emergency braking, even set to the lowest sensitivity levels, are bad to outright dangerous in city (and some country) driving. In NYC it's common to have multiple road lines crossing during construction which by default has lead my car to try steering into cyclists or barriers. It can't read the lane because the lane is not properly marked but that's typical. So I disabled it. Emergency braking is a bit harder for me to understand when it kicks in but I've done whatever I can to keep it off. Twice it caused a hard brake in the middle of traffic with my newborn in the back and almost got me rear-ended. Another time it detected another car ahead of me but technically in the next lane and warned that it was about to hard brake in the rain on the Williamsburg Bridge inner roadway which is... tight to say the least. If there's a future where these features are rigorously tested for worst case scenarios in different driving environments and somehow universally implemented across manufacturers... maybe.
this seems like the sort of thing that would be better measured with actual numbers, not just anecdotes from people on the internet.
the NHTSA is a pretty cautious organization, and they like numbers. if they've studied it and found that automatic braking will prevent 24000 crashes and 360 deaths per year, i'd be inclined to believe them. having the car brake for me when i'm not expecting it sounds kind of scary to me too, but i'd rather experience that than let 360 people die every year.
Eh, that's great for you (according to you), but for the millions of folks that are driving around staring at their phones, I am beyond thankful for this technology.
I've been driving a Honda with the all the safety assists for several years and while it does do the occasional erroneous "you need to brake" pre-warning, I've never had it brake unnecessarily.
Anecdotes aside, I'm fairly certain automatic braking is a net reducer of accidents.
Would you care to elaborate on what went wrong? The other comments here make ‘automatic emergency braking’ sound like a really good idea, and it’s hard for me to imagine that this particular feature would be so problematic.
If you’re talking about other automatic steering features, that’s very different, and your comment makes more sense to me.
I've driven hundreds of thousands of miles in cars of 3 makes (Mazda, Subaru, Nissan) with AEB systems and they never once had a false activation, while the feature did save me from several serious accidents. The Nissan is also nice enough to warn early so I can react before the AEB kicks in.
Properly implemented AEB is an essential safety feature that saves lives.
Fully agree. As it is fine most the time it comes back to the other part where it is wrong.
Example: living in cold weather climate my outback just stops cuts the engine when exhaust from the car leaves car in front of me on cold day( large cloud of vapor). The visual collision avoidance detects this as car and cuts the engine. This is not the place you want to be leaving an intersection with people accelerating behind you.
Unless this backed by lidar / radar or combination that is more accurate I don’t see this as pro. Top reason I can’t stand my outback is the decisions it makes incorrectly.
Of course there is no way to get any of this fixed in my outback other than selling it. Consumers are getting the shaft on technology in their cars imo and are being treated as test pilots.
There appear to be major variations between manufacturers. I've driven many miles on a wide variety of roads in a Subaru and haven't experienced any false positive automatic braking incidents. However, the system does automatically disable itself (and warns the driver accordingly) in heavy precipitation so there is still significant room for improvement.
Most people I know buy older cars now. They're more fun, less cluttered with dumb features and technology. I get it. When I was shopping for a new car recently, all the salesmen talked about was how big the touch screens were. Cool, the manufacturer placed an order for a panel of a certain size, very impressive engineering there.
Touchscreens first, quickly followed by excessively bright, flickering, and aftermarket headlights.
I had an Audi that would emergency brake on the freeway if it was raining and there was a semi in the next lane. Terrifying.
Now I drive another car that will hit the brakes in parking garages when going around corners if cars have backed in and I am going 5mph or more. I've learned to drive 4mph in parking garages.
I just had to junk my 18 year old Buick sedan. Everything sucks now. Even mid range cars reek of the electronics being obsolete or broken in 5 years.
> Now I drive a 27 year old car that has none of these 'features'. It does what I tell it to and when I tell it to, it doesn't phone home, and has no touch screen, but old fashioned buttons.
I second your motion!!!! I was happy with my 1986 Chevy S-10 Blazer until it was badly rusted out by winter road salt.
Now I have a 2001 GMC Yukon. I prefer the Blazer since the Yukon has too many attempts at automation and is too big. But apparently the new cars have so much in automation that they are to be avoided, and cars of the future promise to be worse still.
Agreed here too. Pretty much all my experiences with these kinds of features have been poor. I turned off the emergency breaking on our Tesla after it slammed on the brakes for no reason on the highway a couple of times. Could easily cause someone to rear-end you. I also recall a Kia I rented once that had automatic lane keep assist and blind spot warnings. The blind spot warning was so sensitive that it would loudly beep at you if there was another vehicle anywhere near you in the other lane when you put your turn signal on. Not helpful, and quite distracting. The lane keep assist was worse, because the way it pulled on the wheel any time you approached the edge of a lane (which it turns out you do often in the course of regular driving for all kinds of reasons - another car crowding you, giving room to a cyclist, taking a smoother line through a corner, etc.) felt very much like losing traction. It was quite disconcerting, and I could see it causing someone to overcorrect and lose control. And to paraphrase Matt Farah, if you need assistance to keep in your lane, you shouldn't be driving.
All that said, I do like the autosteer on the Tesla, with the caveat that you need to treat it like cruise control, NOT like "autopilot". And I'll pretty much only use it on divided highways. In other situations its insistence in sticking to the dead center of the lane can be disconcerting at times (like the ones I listed above!)
My son just got a job outside the house. He hated the looks of newer cars and with the current market a Honda Fit with 100,000 miles cost about as much as spent our current fit which was 30,000 when we got it. (Dealer had excuse why they didn’t have any new ones.)
We found a 1996 Buick Park Avenue for less than $5k and keeping it running might be an adventure in maintaining American cars but boy is it a sweet car and it gets 29 mpg on the highway which is not bad at all for a large car.
There are good points here on both sides. I'm in your camp, personally. I'll probably always drive older cars, because that's just my preference, for many reasons. Maybe the answer is to require the feature on new vehicles, but also have the option to disable it. Most drivers probably wouldn't, but those that feel strongly about it would have the choice.
Automakers are also terrible at writing software.
But I do think these safety features are important if implemented correctly. I think the crux of the issue is: when someone _is_ paying attention, safety measures feel like they overreact and get in the way of subtle human reactions to things happening. If the driver is _not_ paying proper attention, then the safety systems are the only thing that react in time.
I feel like this requirement is important, but requires better software, including lessening the impact of safety systems engaging when the drive has been actively steering/accelerating/braking during a period of time, making it obvious that they are indeed paying attention. But there are plenty of drivers who pay almost no attention at all, many people go into auto-mode when driving and aren't ashamed of it.
I've almost been run over by drivers who have been directly looking at me before, then sudden braking when their eyes have stopped glazing over.
My MB van also has similar serious issues with buggy auto-braking. Once, while driving head-on into a bright sun, it decided it was time to apply full brakes while driving highway speeds on a busy highway, in a vehicle with a GVWR of around 9,500 lbs. That wasn't fun and could have easily caused an accident involving multiple cars.
My Tesla, which is probably the leader here, regularly makes mistakes - though never really to the point of forcing braking. Often it will warn me on my narrow street of objects like parked cars it thinks I will hit, but it seems a little more polished than MB.
If regulators want to force this on all car manufacturers, they really need to also mandate certain tests and features - plus OTA updates to recall any bugs... My guess is that most car makers just don't have the workforce to develop these features, so forcing them to roll them out will mean even worse bugs!
I had a very similar problem - a mouse ate through my wiring harness and it was tripping my brakes lights out, which the ECU thought I was braking, and I couldn't accelerate. Took the dealer like 5 trips to figure it out. Insurance covered it.
Btw, where'd you find that old car? Locally or get it shipped to you?
Oh yes, I tried a Honda E for a few months and the emergency brake system was incredibly liberal. I live in a city with lots of parked cars and unusual road dips/bumps. There was a turn into my kid's school road that would emergency brake about 20% of the time whenever I entered it. I almost got rear ended a couple times and it scared the crap out of me.
On the other hand, maybe the benefits outweigh this drawback? Pedestrians get hit all the times by cars where drivers aren't paying attention (looking at their phone, etc.). So for the general population, it is probably a net positive.
Are you talking about emergency braking or lane assist?
We've disabled lane assist on our car as it seems to always take the wrong action, but we've never had an issue with the emergency brake which is the subject of the article.
My wife owns a Nissan Qashqai. One day she complained that car started to brake on doing zip merge and this is dangerous on highways. The root of problem was combined back light with stop signal and blinker. Wires to stop signal and blinker suddenly soldered together and there was some current on stop signal wire on turning blinker on. Car's computer decided that brake switch near braking pedal is broken and started to brake. Fixed by cutting one of wires and using new wire around spoiled point.
Same here, in the last 18 months 2 different VAG cars have hauled the anchors on for no reason at all with me driving. Fortunately, I wasn't in traffic, but both cases would absoulutely have caused an unnecessary crash if there had been a car behind me. This technology will be great when it's ready, but the false positive detection rate is massively higher than would be acceptable.
I'm so glad this is the top comment. Ghost braking is a far worse problem.
I had mine (2018 VW Golf GTI) trigger when driving really fast on a very hilly road. Headed into a dip, the braking triggered at 70mph. Nearly gave me a heart attack.
2 years ago, a brand new Audi (rental) tryed tro chrash me several times into tunnel walls. The car suddenly tryed always a hard right turn.
I have an MB and AEB has been nothing if not extremely useful. It saved me from hitting a pedestrian who I didn't see, thank god.
If suddenly braking causes an accident it's because someone was not following at an appropriate distance. You should always allow space for unforseen hazards. It's actually the law.
If both cars have AEB even that unsafe condition becomes a non-issue.
Just because humans have adjusted to regularly driving in an unsafe and illegal manner does not make it the safest way to use roads.
Also the damn side collision avoidance which is active for way too long post-overtake.
I want measurements like fatalities per 100 million miles driven to guide this conversation because there are so many emotionally driven reactions to Elon Musk and AI/ML that I can’t help but think that comments are motivated by things other than human safety!
That being said, I do think that touch screens are a terrible design decision on almost all devices.
Add to that the ability to stop and rob cars with a roll of foil. Thanks, but no thanks.
Looking at the comments here I need to double check if I'm on some gearhead forum where automatic transmission is considered appropriate exclusively for teenagers and geriatric patients.
Anyway, I'll add my anecdote when my Toyota surprised me with a panic brake in a blind corner which probably saved my life by not ending up under a stopped bus. But the kicker is that my insurer gave me quite a steep discount when I originally bought the car, specifically due to the emergency brake feature. They may have more data comparing collisions for cars with/without this feature than anons on internet forums, who knows?
Wanting to have control over the vehicle is not some "gearhead forum" thing at all. I barely care about cars and I still care about this.
I'm the driver, I should be in control and there's literally no need to justify it any further. Anything that forcibly takes away control is automatically wrong from a basic human dignity perspective. I couldn't care less what governments or insurers think, I'm the goddamn human and I should have total control over this machine. I'm driving, not being driven.
It's not exclusively a "gearhead" opinion to not want a system in your car that can violently take control away from you unexpectedly
Statistics are meaningless to the individual. Sure, emergency braking might be safer across the board, but that means little compared to an individual who already drives safely without it.
Insurance companies are run by people that are good at statistics, and there is a fair chance that statistically this works out well but that does not mean anything for any particular individual. Chances are that EAB will cause some accidents and prevents a whole lot more, and that's enough for the insurers.
Yes everybody here seems to be ignoring that car accidents are the leading cause of deaths for children and young adults… car safety is not a solved problem!
Afaik, insurance rates are heavily influenced by collision data.
This feature once saved me from what would have otherwise been a very nasty crash, many years ago.
I was driving too fast in a borrowed car over a small hill and hadn't realized that traffic had stopped just out of my sight until too late. I heard a beep and saw flashing on my dash and I started braking but a moment later the emergency braking system kicked in, braking much harder than I would have been able to. The car stopped, hard, a foot or two short of the pickup truck in front of me.
As the brakes kicked in, my seatbelt tensioned and pulled me into my seat, which was surprising because, although I had a vague idea of emergency braking, I didn't know the seatbelts would do that. And although I may be mis-remembering since this was so long ago, it feels like the passenger seat next to me started moving (I'm not sure why) – I should look that up.
Made me infinitely grateful for the engineering that went into this system. I was quite young and insufficiently cautious, and it was an eye-opening experience.
Seatbelt tensioners are life savers. They're a little explosive charge under the chair near the seatbelt attachment point that yanks hard on the belt to squeeze you against the chair while the car comes to a stop. They're super powerful and you don't want to mess with them unless you know exactly what you're doing (see also: airbags).
> the emergency braking system kicked in, braking much harder than I would have been able to
So this is completely not true. You can brake as hard as any system can, there's no magic to it. Just press on the brake pedal hard, in any ABS car.
Surprisingly, this is something people often don't do. I am an instructor at car control clinics, one of the exercises is panic braking. All you need to do it press on the brake pedal. Easily 50%+ of students struggle with this. Just press hard. That's it.
Also, please take car control clinic. It will save your life and those around you.
> braking much harder than I would have been able to
Any idea why you weren't able to apply full brakes? I remember reading that most people don't in an emergency and that's why the emergency brake boost was created, but haven't seen the reason for it.
I've had two situations where it prevented me rear-ending someone.
The system also had several misfires, like when there were cars parked on the curb on a long turn, but mine in particular doesn't start braking until I do - it just beeps and makes the dashboard light up in red.
It's pretty reliable and I would be all for mandating at least this basic implementation.
I confess that I find a dark humor in how many self professed "great drivers" find these features a nuisance. All the while so many people are completely unaware that they have these, as they have never activated. Hard for me to square those pegs in a favorable way.
Which, yes, being a great driver means you may be pushing close to limits while still be completely under control a lot. I just don't have a strong prior for that view.
I am reminded of the quote, often attributed to astronaut Frank Borman:
“Superior pilots use their superior judgement to avoid situations that require use of their superior skills”
Nuisance or not is a false dichotomy. Mandating it rather than simply offering it makes cars even more expensive, by making cheap cars even more unprofitable to sell. Imagine if car ownership becomes increasingly unaffordable, like what homeownership has become.
I can't count the number of flowers AEB has prevented me from murdering. But at least in my equiped vehicle, the nuisance activations haven't been dangerous, and most of the events in traffic have been reasonable, if overly conservative. During parking, less so, but at least predictable.
Having AEB without being aware of it is probably not ideal. If it activates improperly and you didn't know it was there, that's twice the disorientation. It was really jarring the first couple times, and if it was on a wet bridge while turning in a narrow lane, that wouldn't be pretty as others have discussed.
Yeah, I have to agree. Maybe I just got lucky with very capable AEB, but I've had it activate on me two or three times - and every one of those it happened basically at the same time as I smashed the brake pedal myself.
i intentionally purchased a car without any of these hanfisted “safety” features that take control of the car away from me. NEVER had an issue. sit close enough so that you can properly slam breaks to engage ABS as hard as automatic braking systems do, and don’t take your eyes off the fucking road. it’s not that complicated, and i’ll be damned if i let a car make that decision for me.
A driver that drives unreasonably more slowly than the traffic that surrounds them is a bad driver. They cause traffic and create more chances for crashes -- after all, crash severity is directly proportional to relative speed between moving bodies.
The only time my vehicle ever used its emergency braking was when my wife’s mom was driving, and my entire family was in the car. I was so grateful for it.
The scariest part was that she accelerated toward a truck that was travelling very slowly in front of us, the car braked so hard we were all pressed into our seatbelts and the car was audibly vibrating as it lurched from the anti lock brakes. We were literally inches from hitting this truck, and her mom was completely unperturbed. Like nothing had happened.
I was shaken. If I’d done that myself I’d think I needed to take a brake from driving and seriously contemplate how that was able to happen. She just kept driving and talking about whatever she was talking about. Disturbing stuff.
So, that’s why I’m generally in favour of emergency braking. It completely saved my ass.
I think this is human nature at play. She knew what happened.
It's just that people don't confront huge mistakes like this head-on.
I don't know how many people have pulled out in front of me, almost killing me and they seem to just wave.
Although once an amazing thing happened. I had gotten up early at o-dark-thirty and was driving through the suburbs and this guy came barelling out of a sidestreet. I hit the brakes and he made the turn, cut me off and drove in front of me. When we got to the next stop sign, he got out of the car and came back towards me. I was about to back up and gtfo, but rolled down the window. He said "it's so early, and I didn't see you there and I'm so sorry." Very nice and I completely relaxed and smiled.
oh, also good time for the old one:
I want to die while asleep like my grandfather,
Not screaming in terror like the passengers in his car.
Sounds like you don't want your wife's mom driving with you and the rest of the family in the car. Sounds like she quite possibly should not be driving at all. That must have been pretty scary.
I see a lot of comments in there in the lines of "it almost caused an accident!". Yeah, my AEB has misfired too, often when coming around a curve and it thinks I'm about to have a head on collision with the oncoming traffic even though we are both turning. Or when making u-turns with a car in front of me.
But it's never caused me to get hit. And more importantly, if the other cars around me also had AEB, then they too would get triggered.
Yeah, it's sometimes annoying when it misfires, but man, that one time it fired in a real situation definitely saved my life as I stopped inches from the car in front of me.
I'd take the minor annoyance of misfires over a crash any day. Especially if everyone has it.
My AEB is overridden by an increase in throttle; even a small one.
If (rarely) it misfires, it’s only pretensioning the brakes initially. That’s just getting the pedal firm and pads against the rotors, no braking force. A small bit of throttle, not even enough to speed up, is enough to tell it to cool it. No actual braking occurs.
Anyone who’s losing control of their vehicle over that is already driving too close to the vehicle’s limits.
But how long do we have to wait for everyone around us to get AEB? 20 years? 30 years?
In meantime, every time it misfires, we’re risking an accident. Yes, it’s an accident caused by the poor following distance of the car behind us, but the blame isn’t much of a consolation when there’s no guarantee you can walk away from the crash without an injury.
There is a lot of concern about incorrect actuations in this thread.
One way to address this is to dictate that the car AEB status prior to each accident was made known to everyone involved, and to hold the manufacturer directly liable whenever AEB contributed to a collision.
After making that change, they could circulate the draft bill, and let the car manufacturer lobbyists kill it.
I don't think that car manufacturers should be held responsible just because AEB contributed to a collision. It isn't that rare for safety features to do harm in specific circumstances, for example seatbelts can prevent people leaving the car in a hurry and airbags can sometimes kill people when they activate.
The question is whether the feature is a net positive, not whether it is a positive in literally every potential collision.
Love the path this comment took.
I was probably lobbyists driving the push to make them mandatory in the first place
Periodic re-testing to keep your license should become mandatory. 50% of federal transportation funding going to transit and related projects should become mandatory. Getting as many cars off the road as possible in the coming years should become mandatory. Reduced speed limits in cities across the country should become mandatory.
https://www.cdc.gov/mmwr/volumes/71/wr/mm7126a1.htm
If pedestrian detection "doesn't work very well" as the linked article* says, we need to take the simple, practical approaches that we know will have an impact, and stop sacrificing lives to the altar of convenience and the "American Dream".
* https://arstechnica.com/cars/2019/10/aaa-finds-your-new-car-...
No way, you and what dictator is getting this passed? No one would get this passed, even the most corrupt possible senate and lobbyists.
The quality of these systems varies wildly IME. If your car has a good one, they are quite useful. If your car has a glitchy one they are dangerous.
How do we keep the glitchy ones off the road? Extended return policy? Make the unit swappable/replaceable?
Set some standards and 100% transparency on cases where they fail, mandatory disclosure of every case where they phantom brake and what the cause and end result was.
Unlike seatbelts or airbags where there is very little downside these can cause perfectly safe situations to turn into nasty accidents.
> How do we keep the glitchy ones off the road? Extended return policy? Make the unit swappable/replaceable?
Treat a failure as a cause for vendor warranty. Either the vendor has to fix it, replace it or refund the car.
I don't ride horses because I don't want my bicycle to have the ability to make bad (or any) decisions mid-ride. Fuck everything about this proposal for identical reasons. I nearly got myself and several other folks killed on the highway in a rental Forrester when lane assist decided it couldn't tell the difference between painted lane markers and pavement seams in the middle of Denver highway construction. If folks don't want to actually operate the vehicle they travel in that's fine, move to an urban area and use mass transit. For myself, I'll just keep driving continuously older vehicles until I finally get around to doing a full EV conversion on a vintage car, thereby permanently sidestepping all manners of similar bullshit.
Most of the comments in favour of this technology seem to be "It saved me from rear-ending someone because I wasn't concentrating". On the other hand comments against this technology are essentially "It had a false positive and almost killed me". As I'm capable of maintaining good following distance, riding to the conditions, braking correctly, and being aware of the traffic around me; I'll choose not to use this technology, thanks.
> I'll choose not to use this technology, thanks.
Roads are public.
I have been rear-ended by another driver in a completely unavoidable situation (for me -- the other driver should've been paying attention; I saw the collision coming and honked but had nowhere to go).
I've also been in an accident as a passenger in a Uber. My Uber driver rear ended someone. I screamed at him twice but by the time he reacted it was too late.
I have never activated the emergency braking on my vehicle.
Never. As someone who works on these systems, the cure is often worse than the disease. Fix the root issue and enforce distracted driving laws, ban in car touchscreens.
I'm becoming an old man at 36... I just want a feedspinning NA engine with a 6 speed on a 3000 lb RWD chassis where everything isn't driver aids.
I know... I know... this isn't practical or safe, but goddamn it, it's the most fun for me!
Sounds like Caterham might have what you need: https://caterhamcars.com/us/models/the-range .
That's great. Keep your fun off the public highway, though. It's too dangerous. There are plenty of racetracks and a race car doesn't need to be road legal or practical so what you describe should be much more affordable.
Risk Compensation theory predicts this technology will increase risky behavior by drivers. It has been well established that people adjust their behavior in response to perceived risk -- and if the car is going to handle the emergency braking it will cause more dangerous speeds and tailgating.
Well first, people respond to perceived risk rather than actual risk, so those paradoxical instances where accidents increase comes when something seems safe - but isn't.
Secondly, consider you have to consider what compensation buys you. People bringing up risk compensation often forget that. Doing a task faster and with less concentration, these are real benefits that makes peoples lives better.
Third, people don't fully compensate for risk. Often a safety technology will both reduce accidents and let you get the job done faster.
I have it on my car. I think your concern isn’t an issue in reality. I have never seen it kick in while being close to someone’s bumper.
It has only kicked in in three situations, all of which have been warranted.
* I was turning onto a main road. Foolishly, I assumed the car turning in front of me went. I started making my turn. It alerted me and avoided a minor, but costly rear end.
* in dense traffic where people cut in and out. In the time it takes to verify a lane change, someone has cut you off or traffic has slowed. It doesn’t not apply heavy braking, but issues an audible alarm and pretensions the brakes.
* once in a situation where I misjudged the braking speed of the car in front of me. The vehicle was a bit smaller than I though, so it appeared further away at speed.
In the latter two cases, the actual emergency braking did not engage. Only the pre warning and tensioning. In both situations, I would have likely avoided an accident without the system, but had to apply by brakes even harder.
My understanding is emergency braking doesn’t necessarily avoid a collision, but it can bleed of speed in the final few feet before an impact. A collision ending 2 feet into your bumper is far safer than one ending 10ft into your car.
Theory is great. What has data shown?
What would really save more lives would be a much stricter license system that is federaly mandated.
What little you have to do to get your license compared to European nations like Switzerland or Germany is absurd. Not only can you drive at 16 already there is basically no test to drive what is basically a tank down the road.
In the north you have the additional hazard of driving in snow and so many have no clue how to deal with it. A "Schleuderkurs" should be manditory in the north so people know what it feels like to lossee control and how to deal with it.
P.S. In Switzerland if you fail your driving test a second time you can't just do it again. You need to get a psychiatric evaluation that you are fit to drive. Yet we still have shit drivers here as well.
+1
Imagine if we were debating right now on if Airplanes should have automatic braking systems due to all the runway incursions ...
Regulatory capture marches forward.
In an environment where small, cheap, runabout electric vehicles could become very popular, I expect lots of "for your safety" reasons why cars have to be complicated and expensive.
Multi-ton vehicles capable of moving at highway speeds are dangerous and kill roughly a hundred people a day in the US[1]. So whether these changes are driven by regulatory capture or not it makes perfect sense to tightly regulate automobiles. Personally I think consumers should be free to decide things like seatbelt and airbag use but if they drive on public roadways they should not be allowed to make cost/benefit tradeoffs that decrease the safety of those around them.
> small, cheap, runabout electric vehicles could become very popular
Agreed that's a promising area. It's possible that the increasing regulatory burden on cars may drive an increase in low-speed vehicles such as neighborhood electric vehicles (NEV) for use cases that do not require more than 25mph/40kph.[2] Since they are exempt from the vast majority of new automobile regulations that makes them comparatively much simpler and more affordable.
[1] https://www.cdc.gov/mmwr/volumes/71/wr/mm7126a1.htm
[2] I am not claiming this works for every suburban commute, just that there exists a nontrivial number of car trips in cities and towns that would be feasible in a NEV.
I mean, I can certainly agree that there is a thing as regulatory capture. I am not clear that this is it, though. Making cars safer has legit improved life expectancy. Anti-lock brakes, seat belts, rear view cameras, blind spot indicators, etc. are all very welcome advances that make modern vehicles much more safe than they would otherwise be.
That's not what is happening here at all. There is a separate set of regulations for those small, cheap, runabout electric vehicles. They are classified as "neighborhood electric vehicles". The NHTSA has not proposed to require automatic emergency braking for them.
Well they have to do something with all the new funding from the latest “infrastructure” bill.
“helped secure a historic 50 percent increase in funding and staffing resources for the agency under the Bipartisan Infrastructure Law”
lmao not every law is "regulatory capture", sometimes laws make dangerous activities that most of us are forced into (like driving) a bit safer as better technology becomes available.
I’m conflicted on this one, it has both saved me, and also almost caused a real crash… it’s save me twice when reversing, once a guy pulled in behind me after I had checked mirrors and started reversing, and the auto brake engaged and saved a collision… but it has also jammed the brakes on when driving down the street because it detected an obstacle that was fixed, and I was driving around it comfortably
If they do this, then
- Make alignment of front facing cameras free (for newer automobiles not retrofits), not a $500 service option. The front facing camera is bound to fall all out of alignment, which makes the system begin to see ghosts. This then causes drivers to simply disable the automatic emergency braking, after 1 or 2 scares. Therefore the system is broken.
- Fine car companies that cause accidents due to emergency braking randomly braking due to ghosts or other spectra, if due to flawed mechanics. There should not be corner cutting in a safety feature. Such fines already exist for airbag regulations.
If we were to mandate automatic emergency braking today with the state of the current technology, it would be a disaster.
I've always wondered why brakes don't have a variable brightness based on how hard the driver is braking. If you've ever been behind someone who has braked hard, part of the danger is not realizing just how fast they're stopping.
Yeah, but really you were just driving close to the person in front. Remember, the person in front could emergency brake at any time. Even if you manage to detect that and emergency brake too, is the person behind you going to be able to?
Just don't drive close. Train yourself on what the proper distance is and stick to it.
Perhaps the reverse lights could activate when extremely hard braking is applied.
New cars they flicker rapidly I thought
An eye tracking feature would be nice if incorporated into the AEB where it could be set to only activate if I’m not looking at the road in front of me. This plus have a disabling trigger on the steering wheel and auto disabled if you punch the gas from a stop. Uhhg… just imagine someone pointing a gun at you from the front of your car to carjack you, you duck while stepping on the gas - only for your tires to squeal and not go anywhere!
From random evasive/tactical driving videos I've seen, you also need to be able to ram attackers, as well as the front wheel of a blocking car with your frame rail. This system would prevent you from using your car as a weapon in lawful self defense.
Holy forking shirtballs - NOOO!!
At least until it is MUCH better. Like 6 orders of magnitude better.
Just watching the "Pre-Collission Warning" red flash and beep on both a Ford Truck and Mazda crossover, there are far tooo many times that it reliably and consistently would brake and CAUSE an accident.
The most common one is where there is a car or obstacle just on the outside of a turn, such as (in the US, left side driver) a left bend in the road with a "Y" intersection going off the right side = outside of the turn, or a vehicle pulling off into a right-side driveway. As I turn left around the bend without much slowing and the other vehicle drives off to the right => "ALARM!!!". It obviously 'thinks' I'm about to rear-end that vehicle, when I know that no such thing will happen (unless I straighten the wheel or turn right for some bad reason). If braking happened here, I'd get rear-ended.
Similarly, I've seen the alert go off on either dirt roads with some random dark spot in the middle, or a regular road with a temporary steel plate over a work trench, or even just a pebble or small branch in the road.
It is clear that all of these miscellaneous items can produce the appearance of an obstacle about to be hit, but that does NOT mean that braking is appropriate.
It looks like instead of moving forward to an all-electric vehicle, I may be going back to restore some early 1990's vehicles and run them until the petrol market is no longer sufficiently viable to support common corner petrol stations.
Just a few hours ago, I saw an pristine but old Porsche 911 with antique plates. Obviously still way fun to actually DRIVE. Seems like a good way to go!
So in some ways it's hard to argue against new "safety" features; but on the other hand there is a cost to sprawling feature creep in cars (or software): increased complexity surface; increased failure modes; increased repair costs. How many sensors/computers are on new cars; 10 years out how many of them will start to fail/go flaky. Right now I apparently have an O2 sensor and a crankshaft position sensor failure on my vehicle; they do not seem to be impacting normal operation, but replacement of them would cost $600
Yeah, well put. There are a lot of difficult to quantify cost for new features:
* monetary for the consumer (new features give manufactures an excuse to keep the price high) * environmental (because increased failure surface area causes consumers to need more repairs/upgrades) * safety (side airbags hurt visibility... Bigger cars are more dangerous ... generally safety features may make more people drive more dangerously [1]) * increased oligopoly (complex safety requirements stifle competition)
The problem isn't that there's no benefit, it's just that all of those "unintended" costs are actually boons to the manufacturers, so as long as they're calling the shots new features will just keep on rolling. Meanwhile, we just get more and more dependent on something that fewer and fewer really understand.
That o2 sensor is an important part of the engine/emissions management system. Without that info, the car computer is not able to efficiently map the fuel/air mixture and over time is likely to lead to more problems. Depending on the car, the CPS sensor could be really critical (like if you've got variable valve timing in an interference engine I think?). Understandable you want to ignore that check engine light but recognize that $600 cost to repair could balloon into a lot more (even a destroyed engine).
Traffic safety would be greatly improved by a system whereby cars could automatically communicate their speed, rate of acceleration / deceleration etc. to each other. Has anyone considered using brake and parking lights as a way of doing this? I could imagine that by rapidly flipping on / off some of the LEDs, it would be possible to communicate simple information to following vehicles with capability to read the signal, while to other vehicles without such capability they would flicker so fast as to just look like normal brake lights.
Yet another classic example of perverse incentives. Create a problem, offer a solution that creates other more serious problems. Force it through bureaucratic circles without representation. Create a less free world.
When you drive, you are responsible for you and your passengers safety.
How can you possibly be responsible for the car suddenly stopping when you have no control over it as a mandated safety feature.
The only choice you have if this is mandated is you can't drive. So you can't travel long distances, or transport things long distance, interferes with interstate commerce and travel you are confined to your local area of walking distance.
This is beyond stupid, and lacks any basis of granted authority since it opens the door for violating constitutional rights, in the name of a dubious claimed safety issue. Congress can't create a law and have this not be unconstitutional how can a committee think this will be any different.
Given how many people drive cars every year in the US, 360 lives saved is a rounding error. Would like to see the data on the accidents it actually prevents. I have cars with and without this. Honestly I haven't noticed a difference one way or the other with anecdotal experiences.
Perhaps I'm getting jaded (or wiser) in my old age, but anytime I see "mandate", I just picture congress people shoveling truck loads of money into a company's stock. So if this is mandated, what stock should I buy? (I'm serious).
Mandated safety features wouldn't increase stock price of any of the auto manufacturers. You might be able to expose yourself to the increase in price if you can isolate the manufacturer(s) of the technology involved itself. You're still highly unlikely to capture any alpha if the various car companies diversify which chips/camera's they are using.
So we're completely giving up on people being in control and responsible for these things? Fine. But this is a stupid reaction to that. Cars are far too heavy, far too powerful, far too comfortable and far too distracting. They are too much of a good thing. We desperately need to stop these things. But we're addicted. We're destroying the environment, hurting ourselves and others with these things. But we can't stop. Just like a heroin addict.
Reading this thread, I can't help but wonder if some folks are confusing phantom braking on radar-adaptive cruise (a really widespread issue and common trait of radar based cruise) with AEB (Automatic Emergency Braking) activation. In my experience the former is common and can feel scary at highway speed, but this is not the same thing as AEB, which is a separate system. My ford for example has AEB and no cruise control at all, and has never once misfired.
I have had more problems with such driving tech than it has ever helped. I wonder how the data to justify this is collected.
A while back I saw this article about a Tesla that stopped on a highway and caused a massive pile-up
https://theintercept.com/2023/01/10/tesla-crash-footage-auto...
Hmm, perhaps you could also mandate better braking systems in trains while you're at it. :-)
Can these systems be FOSS? For example, if my car would have such a so-called safety feature, I have a strong feeling that I must know all possible incomes and outcomes, for understanding what has fired the emergency braking if it fires somehow.
Can confirm phantom braking is a problem. I was driving an ionic 5 and it activated emergency brakes despite not having any forward sensors (due to an earlier accident) and braking supposedly being disabled due to lack of sensors.
Next it will be, "watch 30 seconds of this ad to unlock your brakes."
That gives a whole new meaning to 'anti-lock braking'.
Bought a VW and a BMW with these features in the last 6 months. Both can disable them but we find them to be worth leaving on even though the occasional false positive can be jarring.
Funny, when will the feds mandate a serious driver test...
I have a sense that my Model Y has gotten better — less phantom breaking. Curious what other people’s experience with Tesla have been lately…
sadly yes they are right. the implementations are sadly lacking so as a competent and always aware driver, i’ve found that 100% of the time these systems (on all levels of car, honda to mercedes) these systems do more harm than good. however your typical usa driver is quite poor, so yes these systems need to be required
I wanna know the quality of software developer engineering these features. Are they hiring bootcamp grads?
This might be controversial, but I don’t want any automated systems. I want a mechanical car and I don’t trust government or tech. Particularly in the privacy space primarily.
I say this as someone who spends hrs a day on a computer building AI systems…
I have a modern car (2019) and the auto braking, collision and lane drift alert is a nuisance. False positives are common and there is no way to disable most of it. Some of it can be turned off but as soon as the car is restarted it resets to the default. I've even had the car emergency brake and come to a complete stop in the middle of the road due to a phantom collision. As soon as I pay it off I am going to sell it and roll back to an earlier model without all this mandated, modern tech and keep the older car rolling till I die. We will become like Cuba keeping 50's cars running into the 70's or even the 80's and this time it won't be due to the commies but the fascists.
Name the car brand and model. Not all implementations are so bad, so it’s important to raise awareness of the bad ones.
Weird that you cannot disable the collision warning and lane keeping. What Make/model is it?
As soon at the becomes prevalent, people are going to start fucking with it. For instance, I'm personally going to bring a big piece of posterboard with a car printed on it, and hold it out in intersections that I'm about to cross on foot.
A feature that lets an adversary take control of your car in this way is known as a bug.
A high-visibility vest worn while walking on the side of a road is sometimes all that it takes to activate it. :)
I have never understood how this would work in the real world. If my car automatically brakes, I would just get cutoff by other drivers constantly until my car comes to a complete stop in the roadway.
And yet I have it in my very real world (and unexciting) Subaru for years and it never does what you say. I’m guessing it has some idea of differential speeds , not just distance-to-object
You're thinking of adaptive cruise control. Automatic breaking only activates when you're about to crash.
Different cars have different systems, but mine has never triggered for a car coming into my lane. Sometimes it's triggered on cars leaving the lane while slowing though.
Otoh, I won't use the adaptive cruise control, because the granularity for following distance settings is too coarse, so I have to pick either follow closer than I'd like or follow far enough back that people treat it as an engraved invitation.
I'm not a big user of cruise control anyway, so no big deal. I'm also less frequently driving long distances on I-5, so not in that situation much anyway.
The point is to gradually remove private automobiles from the transit system by incrementally nudging out their availability, either through expense or added bureaucratic hassle.
The WEF has discussed the need for this repeatedly: don't forget, you'll own nothing and be happy AND safe!
My new car has lane keep assist and AEB and I haven't had these issues. From what I hear it's very much to do with a maker's implementation (mine is Hyundai, who for various reasons seems to have a bad reputation in North America but a good one in Australia). For the lane keep assist I feel a nudge but it's extremely easy to overcome just through a normal grip on the wheel - if I don't budge it won't move. It's been an issue near construction sometimes but also some wider merges where 2 lanes become one (the markings in Canberra seem to confuse it, but not Sydney) so it does depends a lot on road conditions, but as I say it's annoying at most and I wouldn't call it dangerous in my car and it helps more than it hinders.
For AEB, the same system is used in the adaptive cruise control and it's been near perfect. One funny thing was that it just alarmed and fully died in outback Victoria - apparently it is coded to do this as a fail-safe if it hasn't detected a car for a long time due to the system potentially not working. The service centre employee said Hyundai are working on a patch to extend the timeout since it's a little low for country Australian roads (approximately an hour at the moment I think).
> From what I hear it's very much to do with a maker's implementation
Right, if I had to condense my issues into something more constructive it is just that: if regulators want to pursue a mandate for these systems, they should also work on establishing testing criteria and specifications that cover the many different conditions drivers face and reduces variation between manufacturers as possible. Not an easy task.
To give a quick example: another commenter in this thread states that lane assist only kicks in at 35mph. That is true in some models but not all which could lead to some nice surprises. Further, the speed limit in NYC is 25mph so it's not unusual if the road surface is decent to be cruising closer to 30. Unless there's a threshold for activation, you might be walking a fine line between "safety on"/"safety off" and any resulting corrections it might make.
I have a Mazda too, I can confirm GP statement regarding lane assist... in many areas but not all:
- when it works it works so well the car basically drives itself esp. when you throw cruise control into the mix (this includes auto coasting + breaking)
- when it doesn't it's bouncing across the lane + applies too much resistance when attempting e.g overtaking
I did not have any issue with emergency braking (never triggered, any situation where it would have triggered got handled by the driver first), but I did have a few (light) phantom braking events in cruise control.
Not related to lane assist control (uses a separate mechanism) but I did have issues with blind spot alert, it was randomly alerting regularly when no car was present. Mazda identified the cause as winter tyres throwing the sensors off. A software update is supposed to fix that but we disabled the feature so it'll be done on the next scheduled maintenance.
Maybe also it depends on the type of traffic and local driving habits and what saves your life in one city or country could kill you in another. The thing is, I’d rather have it disabled and not save lives but also not take any by default. It will surely improve over time and if it becomes flawless sure, I’d take it. The thing is that I don’t wan’t to be beta testing with my life or my passengers lives.
In cases where it’s not possible to disable the feature we better constantly record the car from all angles to have it as a proof against the manufacturer in case something goes bad. Near accidents caused by buggy software should not get a free pass, but require the manufactuter to fully reimburse you or pay you some serious fine.
Hyundai had a not so great reputation in the early 2000s in the US. In recent years however this opinion seems to have shifted.
Been driving an Ioniq 5 2023 for a couple months now with no phantom/emergency braking issues. I'm not in love with the performance of the lane keep without being on an HDA2 marked highway. I have to fight it just enough for it to not be useful on certain well marked backroads.
Considering tossing in a Comma 3 to give no-hands driver assist a shot.
Yes, *near* perfect is the correct word. Anecdotally, in a 2021 Hyundai, I do get random AEB warnings in the city mostly when merging into fast-flowing traffic.
Thankfully it is possible to have the system set to beep only as slamming on the brakes while merging is an experience everyone wants to avoid.
Also, it sometimes panics when coming to a stop on an incline with a car ahead. I guess it is unable to account for the fact that the car will stop by itself just because the road is going uphill?
Driving a Hyundai hire in Sydney with lane keep and aeb was the thing that eventually convinced me to upgrade the 4x4 to something with these features. Driving a hire Subaru with the same features around Sydney was what convinced me to drive a few more and observe who had good implementations, and who poor.
Recently upgraded to a Chinese ute. I've about the first 1000km driven now, and whilst it's occasionally made some mistakes, if my hands are on the wheel it's easy enough to just stay the course and it let's me do as I wish. I can also turn all the setting to off at the touch of a button and make that the default too. I quite enjoy having these features because they have been implemented well.
Bad North American repurtation comes from dealer network afaik.
If you haven't already done so, please file a formal safety defect complaint with the NHTSA with details on that bridge incident. Once the regulators receive enough similar complaints they will open a formal investigation and potentially issue a recall.
Good one, not in the USA but I will do this here in NL with the equivalent service. It never crossed my mind so thank you very much for this option.
Bug reports filed through NHTSA?
Good idea, thanks for the link!
I also had a very similar experience with lane assist trying to send me straight into a cyclist. Had a newish rental car with lane assist for a day and didn’t even really notice I had lane assist until driving out on a country road in California. Cyclist was up ahead staying nicely in the right of the lane and no one was around so I was giving him plenty of room driving to the left until the lane assist kicked in and pulled me rapidly almost straight into him. If I didn’t have a firm grip of the steering wheel it would have been much more dangerous. I feel bad for him as he was probably fearing for his life and thinking I was a horrible driver…
I had AEB trigger going through a mini-roundabout in rainy weather as another vehicle was yielding to me. I also had an AEB alert when fresh snow was flying off the front grill just after I started driving for the day. This is on a 2021 Sorento.
Oh, the Mazda lane assist and lateral collision thing from the side mirrors I found very weird. We rented a MX-30 at one point and were driving on the highway, used the signal to change lanes and mysteriously it decided that there is a car on the other lane (there wasn't) - warning in side mirror, and nudged the wheel and refused to change lanes. It was so weird having the car not allow me to do something because of a "ghost" it saw. Even with our current car, which is not a Mazda, had 2 small incidents in 3 years. One at night while it was raining the front collision warning started beeping, no break though (nothing in front, except rain) and then again, which was a bit dangerous, on the highway on a sunny day, no one in front, somehow I suspect the camera system thought the shadow of an overpass was a car maybe and decided to break all of the sudden (with ACC on), luckily there was no one behind us, but still, was a bit scarry.
> somehow universally implemented across manufacturers... maybe.
They'd need to be implemented and then turned off for 15 years until 99% of the old cars are off the road and then flipped on simultaneously for every car.
If a car suddenly breaks then the car behind it better be able to suddenly break too.
The need to suddenly brake is not a new phenomenon. Cars without AEB can achieve it the old fashioned way. The driver applying the brake.
That sounds like there is a lot of commonality between the two cases that I had and the ones that you have experienced. My car was a Mercedes Benz C class July 2014 issue. The AEB relies on a lot of gear supplied by Bosch (I dug into it a bit to see if it could be disabled without surgery but that wasn't possible). Maybe Mazda has the same unit? (it sits behind the front license plate and there is a camera behind the windscreen).
I'm not sure of the exact sensor. There is a little more info on it here for example (https://www.mazdausa.com/static/manuals/2021/mazda3/contents...) but I can confirm the sensor placement is the same behind the windscreen, the lower one I think is by the badge. I wish they'd just made dashcam a standard feature instead and forget about trying to throw me through the windshield for shits and giggles.
> I wrote up a long comment that got scrapped by my focus filter
I've been making productive use of the "textarea cache" extension lately. Sites and browsers lose text a lot more often than they should.
The B in AEB stands for “braking”, not “steering”, so I don’t understand what system you’re referring to that would steer into cyclist traffic
Lane keeping assistance, from the context.
>focus filter
I’m more interested in this
Just https://heyfocus.com which I must have installed through Setapp, as I don't remember paying for it. It has a scripting capability so in theory I might be able to do something to prevent it from triggering when I'm active in a form.
The problem with numbers is that they don't tell the whole story.
I can easily imagine a scenario where it might prevent 25000 crashes a year that are caused because 5% of drivers are staring at their phones, have a high chance of rear-ending someone, and this software makes that far less likely. While also containing bugs causing 1000 crashes a year for the larger number of non-distracted drivers, resulting in a net change of 24000. This would mean that adding the software would be better for society, while making the majority of drivers slightly more likely to crash into someone.
No, that isn't a problem with numbers. That is a problem with letting your imagination run things.
The scenarios and questions you ask are easy to explore in the data. If the number of crashes goes down, contrived scenarios don't really impact things. If they stay flat or go up, then you have to answer why.
To emphasize, for your imagination here to be relevant, the number has to do one of those. If the number goes down, even if it shifted some situations into a wreck that would not have been before, it must have done fewer of those than it moved wrecks to non-wrecks. If it somehow causes more situations to be wrecks, the number, by definition, goes up.
That's true of basically any change isn't it? Even seat belts conceivably could result in some very small increase in the chance of death as a result of it being harder to escape a vehicle. Obviously the benefits far outweigh the risk, but then we are just arguing over where to draw the line.
As someone who does ML for a living I can definitely say that false positive data from the field can significantly improve the experience for those 1000. That is what should be standardized if we wanted to see systems that work for everyone.
I’d imagine this is exactly the sort of analysis the powers NHTSA’s recommendations. They collect huge amounts of data on this stuff.
> the NHTSA is a pretty cautious organization, and they like numbers.
They may be cautious but they're not necessarily smart. Their vehicle safety ratings only measure the safety of occupants in that vehicle. Result? An arms race where cars keep getting bigger, heavier, and taller to do better in crash tests against bigger, heavier, and taller cars. These "safer" vehicles are now more deadly to everyone else, especially pedestrians and bicyclists.
McNamara fallacy in action.
And the actual numbers should include the opportunity cost (in lives saved or QALYs, if you like) of the increased price of vehicles with these systems. Could that money have been spent on smoke detectors or healthier food or ?
National Highway Traffic Safety Administration
I agree.
I also agree with this:
> If the government wants to immediately improve car safety they should outlaw touchscreens instead.
i also hate touchscreens in cars, but that's not an either/or situation with AEB
It sounds like they should do both, then?
> but i'd rather experience that
I experienced it multiple times and not even once it triggered in a correct situation. Usually a weird angle on a t-junction triggers it unnecessarily.
But they didn't study anything real world. They looked at all rear-end collisions, assumed that a percentage would be prevented and did the math to result in the 24000/360 numbers. No analysis of other affects (like sudden stops causing crashes from the cars behind them). No extrapolating from measured data.
> No analysis of other affects (like sudden stops causing crashes from the cars behind them).
At least part of that effect will show up in these vehicles being rear-ended themselves (as they're exhibiting that exact sudden-stop behavior), and that sort of accident should be reduced if all cars come with AEB.
You'll never be able to wipe out all variables in this sort of study, but I don't think the methodology's as flawed as you're asserting.
360 deaths is noise.
Exactly. It's 0.92% of the 38,000 car crash deaths in the US annually. I agree, this is noise and honestly it doesn't make sense the NHTSA would even consider a mandatory "safety" feature like this for that small of an expected change. Personally, I will not be purchasing a car with this feature until I'm forced to.
Was that determined by looking at the statistics or is this just your gut-reaction?
I agree, but... false positive automatic braking can cause severe injury, too, though. And I worry that that is just harder to measure.
The article says:
> A metanalysis in 2015 found a 38 percent decrease in collisions for vehicles that utilized such a system, for instance.
Barring massive fuckups in methodology, that should account for false positives that cause an accident. The number of non-collision severe injuries (are we theorizing folks get their head banged against something when it brakes heavily?) is likely to be pretty small versus that 38% stat.
May I ask why the respnsibility falls on the car instead of the driver on the phone in the first paragraph?
Just like you are responsible if you used a gun, you are responsible on how you drive a car.
Because, like guns, it's obvious that the despite the fact that the actor _should_ be responsible, the consequences for failure to do so fall on everyone else.
If my car has been destroyed by the driver behind me because they were looking at their phone, it's little consolation that they shouldn't have.
Not the person you were responding to, but here's an anecdatum:
One time I was driving behind someone under otherwise normal conditions, and they turned right into a parking lot. As they turned right, they obviously began exiting the road. The screen on my car screamed "BRAKE!" and then the braking system activated. They were about 60% of the way into the parking lot, and continuing to exit. I was at a normal following distance, going a normal speed, and they were just making a right into a parking lot. This was a very ordinary situation, and the emergency warning was unexpected.
Now, technically, if that car had completely stopped for some reason instead of continuing to exit, I guess I could have hit the back corner of the bumper. That is, if I hadn't, you know, shifted a couple feet to the left to avoid it.
Ironically, the emergency braking alarm was so distracting that it might have caused me to hit the car in front of me if the brakes hadn't activated, and if I had needed to steer to avoid an accident.
And if the person behind me hadn't seen my brakes suddenly activate, they might have hit me. Or, if the person behind them hadn't seen their brakes activate, they might have hit them. And so on.
I've had this exact situation happen on numerous occasions. It's always people turning into a parking lot ahead of me, in a situation where they will absolutely be out of the way long before I reach them, but also I'm actively shifting to the side to go clearly around them anyway. In none of the situations was there any chance of an accident even if the turning car stopped suddenly instead of completing its turn. And this has happened to me on multiple different vehicles.
Each time it happens my immediate reaction is "what the hell is wrong with my car" before I see the "BRAKE" on the dash or heads up display and realize this happened yet again. I'm surprised I haven't been rear ended on account of it yet.
I fully understand that frustration, but I would point out that continuing through the departure zone like that at full speed is prone to accidents with another car entering the road in that blind spot and you should adjust accordingly. Additionally, following the car in front of you so close that you hit them if they brake suddenly is inherently dangerous and a bad habit as well. Interestingly enough, a universal AEB requirement on all vehicles would help in that scenario.
To be honest it sounds like you're making a damned good argument in favor of mandatory AEB. You're risking a collision because you assume that the car that's turning will behave in the way you expect it to behave (no judgement, I usually do it too). If you're wrong there's a high risk of a multi-car accident because naturally all the drivers in line behind you are tailgating.
I've had this happen too but after the first few times I understand why it's doing it and try not to follow too closely in these situations.
I've detailed this in older comments, in a nutshell the first time was on a wet bridge where the car mistook the return of the bridge supports for an imminent crash causing the car to go into a spin (and on a narrow bridge that could have ended really badly, especially because of those supports). I had the car checked out and it was declared healthy, and that it must have been something spurious. A short while later it did it again, this time in a corner where an advertising sign was placed such that oncoming traffic could read it, again the car braked violently resulting causing some loss of control so it swerved into the oncoming lane. Fortunately no traffic there. I sold it right after and won't be buying another MB, ever. They've really lost the plot if they believe that this sort of thing is acceptable in a production vehicle for their regular clientele.
> The other comments here make ‘automatic emergency braking’ sound like a really good idea
That's because those other comments all assume that "automatic emergency braking" will only do exactly what it's supposed to do and will never make a mistake. But any sensible person knows that's not going to happen. These systems will make mistakes, and some of those mistakes will cause harm. So you can't just make a blanket claim that such systems are good. You have to actually look at the data and balance the harm they avoid by doing what they're supposed to do when they're supposed to do it, against the harm they cause when they make mistakes.
The obvious next question is, does the proposed rulemaking here consider the consequences of the system making mistakes? As far as I can tell, the answer is no, with just one qualifier. The claims about crashes prevented and lives saved are not net savings, accounting for harm done by the system making mistakes; those claimed numbers assume the systems will always perform perfectly and will never cause any harm, only prevent it. The one qualifier is that the proposal does include two "false positive" tests to try to spot systems that brake when they're not supposed to. But there is no analysis given to show why those particular false positive tests were picked or why they would justify the extremely high level of confidence in the systems not making errors that the rest of the proposal shows.
I think there is one other aspect to consider. Over here at least, generally if someone rear ends you it's always their fault. The law says they were not maintaining a safe following distance. So from the perspective of the manufacturer they would probably say such mistakes should not by themselves cause problems unless someone else is breaking the law, no different to a human tapping the brakes because they thought they saw a kid about to run on the road but it was just a jacket stuck on a tree.
You should be safe and free to brake without deep introspective thought, without fear of causing an accident. The real world is not that simple but that's how the law is generally setup.
And as a further point they'd probably say that if that person following had such a system, there's also less chance they'd rear-end you in turn. I was wary of the systems in my own car, but after 6 months of driving they are far more reliable than I'd feared and I'm quite impressed.
This is a solved engineering problem domain: use redundancy. There's a reason air travel is safe despite using many intricate safety technologies.
Not the poster
Vw golf r 2016 the sensor was low to the ground and it would randomly brake for stuff it thought was a threat like a cardboard box.
Subaru crosstrek 2018 - would brake for no reason, cars turning up ahead of me, etc.
VW atlas 2019 - brake for no reason would also scream at me and show a red symbol on the hud when nothing was in the road in front of me.
Subaru impreza 2023 - brakes for turning vehicles up the road, had it brake in a car wash.
On my vw vehicles i bought a can bus programmer so i could turn off as much if the safety stuff as possible because it made the car less safe to drive. In the impreza (current car) turning off the eyesight features is part of the pre-drive ritual.
I also had the lane assist on the crosstrek almost cause an accident on the highway at speed. Roads here are inconsistent and it was violently jerking the wheel trying to stay between lines.
> Subaru crosstrek 2018 - would brake for no reason, cars turning up ahead of me, etc.
Another anecdote - I have the exact same car and haven't had it do anything unexpected. It would really help if Subaru et. al. had ways to offer feedback back to them, because experiences like yours should be shared with them so they can understand what kind of corner cases some drivers are getting into.
> I also had the lane assist on the crosstrek almost cause an accident on the highway at speed. Roads here are inconsistent and it was violently jerking the wheel trying to stay between lines.
I've also not experienced this - the lane assist is just a nudge and if it's overpowering you you should take it to the dealership, this is not normal.
There is a hardware button on Subarus to disable EyeSight when you go through the car wash.
Anecdotal, but the alarm before the auto braking in my Crosstrek (2020) saved me two times. And I was in the car with a friend who activated the auto braking when they had just purchased it and weren't used to driving it.
Why not get a car without that stuff?
Anecdotal, but I was riding with my friend on the highway (~70 mph) with absolutely nobody in front of us, and the automatic braking kicked in out of nowhere. Thankfully the one other person on the road that was a bit behind us swerved and avoided rear ending us.
After that experience, I don't want a vehicle with that functionality in it, and don't want other cars on the road having it. It's half-baked at best.
It had happened enough times to him, and he complained to the dealership enough, that they bought the vehicle back from him. Who knows if these incidents were ever reported to the NHTSA since no physical collisions with other cars occurred.
Make and model?
I’ve had the automatic breaking in my Jeep take effect needlessly, which can be quite jarring
My Kia has AEB, Lane Keeping, and Adaptive Cruise Control. They have false positive type events from time to time, but generally you just know when it's going to happen. The first couple of times it was a shock, but I've never had it trigger in such a way that I felt like it would cause an accident. That being said, you can't just leave it un attended. You're still the PiC of the vehicle.
What it keeps doing to my wife is stopping the car (Audi) in oncoming traffic when she is making a left turn from a median across a road that is two lanes in each direction. She is basically jamming the accelerator to go, and so it applies the brakes for a very hard stop, which once resulted in the car behind her not being able to stop.
Amusingly a poster above indicated they had issues with multiple Subarus, and another had issues with Hondas (which is what has worked well for me).
People on this forum want it to be a technology issue, and I'm sure it is to some extent, but I think the differences between driving styles is also a large component. If you're a careful driver that usually leaves a lot of separation, many AEB implementations will rarely trigger and you'll usually understand why when it does. Whereas if you're an aggressive driver that cuts between vehicles, leaves <1 s separation, etc., you will notice a lot of "unexpected" braking from these systems. Like the poster above that was upset the system slowed them down when they were nearly clipping a car exiting the road.
> the system does automatically disable itself (and warns the driver accordingly) in heavy precipitation
That would have saved me one of two incidents so Subaru may well be the one doing this right.
My parents have a couple of newer Outbacks. My father, who's rather technical and pays close attention to these things, had commented repeatedly that he likes how Subaru appears to be more conservative in deploying new technology, but what they do deploy seems to be functionally solid and correct. (His opinion; YMMV.)
Also, they still provide physical knobs and switches for the important things; that includes climate controls and, I believe, the stereo.
I'm driving an 11 year old Forester and just as happy not to be dealing with all this crap quite yet. And the various satellite connectivity options on it have been left never activated.
Anecdotally, I’ve been driving a 2019 Forester for the past few years, and every year or so - I’ve had an experience where taking an exit off the highway, it detects an object that isn’t there, and starts beeping and throwing up red warnings.
From what I can recall, it’s never gotten to the point of engaging the brake, but it’s super unsettling every time it happens.
When shopping for a new car Subaru was very proud of its dual-camera setup, I don't know if it's actually better but they make a lot of noise about it. Maybe there is some truth to it.
Do you have experience with non-tesla autobrake? Doesn't Tesla have a bunch of "We don't need radars, cameras only" stuff holding back their capability?
>they really need to also mandate certain tests and features - plus OTA updates to recall any bugs
OTA updates for braking systems will result in hackers wreaking havoc. The security implications of this are enormous; it's like putting your country's nuclear reactors on the internet.
Updates should only be done in-person at the dealership, for free.
They would need to mandate certain security requirements then.
So the rule needs a rule to fix it, which needs another rule to fix that, and so on.
Locally.
How would you stop a car with a roll of foil?
I'm baffled by this comment. For one thing, this is literally not how vehicle regulation has worked in any western country for at least half a century, probably longer. There are already legally mandated systems that take away driver control in certain circumstances, notably ABS.
For another, the NPRM being discussed assumes that essentially 100% of new vehicles will eventually come with AEB systems in the near future. The proposed rulemaking is just formalizing that situation and setting performance standards.
Huh? ABS doesn't take away my control, it improves my control by preventing conditions that would reduce traction. This system is always engaged by brake pedal input from my foot. It's not something that cars just do out of nowhere.
The emergency braking system will automatically stop the car without your input whenever it wants. Judging from this thread, that could be due to anything. False positives have been reported. People should absolutely be able to turn this thing off if they don't trust it.
ABS does what I tell it I want to do, giving me greater control of the vehicle. When I press the brake all the way down, it means I want to stop as fast as possible, which I can't easily do without ABS. AEB does whatever it wants, without any input from me, leading to "surprises". ABS is user engaged, AEB takes control away from the user. Stop being disingenuous.
I strongly disagree. You are not the only person on the road, you are creating risk for other drivers and pedestrians every time you drive.
Vehicle regulation should be based on the overall safest option, not 'dignity' or freedom or individualism anything like that.
> You are not the only person on the road, you are creating risk for other drivers and pedestrians every time you drive.
Aren't you notice, my dear safety fan, that the thing you are propagating is effectively adding one more person/actor on the road - a programmer of potentially buggy software, totally not FOSS and without any responsibility for any possible outcomes of possible bugs?
So what? If that's the price to pay to maintain control, fine.
These precedents are also creating risk by the way. It hasn't been that long ago since news was posted here of security researchers remotely cutting off car engines or cars repossessing themselves if you don't pay on time. That's the sort of existential risk you're paving the way for by accepting "safety" in exchange for control. It's not really your car anymore, it's the government's car, the insurer's car. What other policies would they like to set, I wonder?
This is a fine attitude for a closed course racetrack, but not appropriate on public roads given natural and universal fallibility.
Fallibility is not a valid reason to remove control from humans. A human should always have command over the vehicle, no matter what. You could argue that some humans shouldn't be driving cars at all and I'd agree.
> Statistics are meaningless to the individual.
This is just blatantly untrue.
> means little compared to an individual who already drives safely without it
Until emergency breaking stops a bad driver from T-boning you in an intersection as they run a red light.
“I’ll just drive safely” was the argument against seatbelts, airbags, and antilock brakes. Motorcycle helmets, too.
>> Statistics are meaningless to the individual.
>This is just blatantly untrue.
I mean, it's not. Statistics describe how society on average acts, not how you personally act. Someone telling you "Statistically you're more likely to drown if you own a pool" doesn't mean as much if you bought a pool and then became a paraplegic, for example.
Saying "emergency braking will bring down average car deaths" is probably true, but that might not make a difference for Joebob who drives his truck down empty County Road 230 once a week to check his mail.
Forward emergency braking systems don't detect such situations. You'd be hit regardless. They only have a 170 degree field of view at the best of times in their current implementations. Somebody two hundred feet away closing at sixty miles an hour will hit you in the side regardless of forward emergency braking simply because it's not designed to look for something moving that quickly that far away at that extreme of an angle.
That said there's also a balance in safety systems between personal responsibility and systems responsibility. Relying too much on systems responsibility actually causes more harm because the person becomes more careless or is entirely at the whim of the system responsible for their safety. We put railings on stairs in public places but don't require everyone to put on a safety harness and latch it onto the guide to climb stairs for example. There's an expectation that system safety has hit a reasonable limit and the safety harness is too costly in time, money, and attention to implement.
I realize you already had some discussions on this point, but perhaps I can shed some light — I think your disagreements come down to a difference in interpretation of language.
Here's one interpretation of "statistics are meaningless to the individual": If X makes people in general statistically more likely to do Y (such as "avoid accidents", in this case), it does not say anything about the effect it will have on a particular individual's behavior, such as you or me. In fact, it could very well make you or me less likely to do Y. As long as it averages out over the population, the statistic is still true. But you need to know the individual in order to know how X specifically affects them.
I think you are taking a different/broader-scope interpretation of the statement. Something like: if X makes people in general (...as above), then given that we live in a society of various people, it will have some at least some impact, perhaps indirect, on all of us (c.f. your comments about taxes, insurance rates, etc).
Maybe you got that already, but I'm a sucker for helping people be aware of when they're talking past each other (:
> > Statistics are meaningless to the individual.
> This is just blatantly untrue.
This statement is blatently untrue. If a "driver gets in an accident every X miles/km" does that mean if you drive more than X miles, you're "overdue?" No, that's a statistical fallacy. Aggregate statistics are only meaningful if you're within a standard deviation of the mean. If you are a safe driver, it's likely to be meaningless. If you are a reckless driver, you're unlikely to live long enough to find out.
Can current emergency braking systems realistically stop a t-bone accident?
Sure, if it was equivalent in capability to a class 5 self driving car, it would see the light was red, and notice the partially occluded other car, then stop in time.
But, with just a front facing radar? I don’t see how it would be able to do much. It would only have a few feet to stop.
I have two vehicles with this system and they have never “violently taken control”. In most situations, the system gives an audible alert while pretension the brakes. If you react within the alert window you retain full control with the benefit of firm brakes (y’all ever remember getting soft brakes?).
I don’t even want to know what situations I’d need to be in for my vehicle to actually emergency brake, but I don’t want to find out. It seems like I’m already going to be hitting something anyways so any speed it can bleed off is better.
Same here: Honda CR-V 2020. It uses single lens (so not a stereoscopic vision) and radar. Never did it brake hard, only warning and few times brake pretension. It will occasionally give false positives on a street curving left with cars parked on the right side. Once it did warn on a highway most likely due to metal comb joint of the elevated road section. (I specifically looked at what could be the cause) The driver reaction within a time window is likely important factor in preventing the system to elevate. I have good reflexes and will put my feet slightly on the brake in such situations.
Honda had a recall for an older AEB system 2017-2019 Accord and CRV:
https://www.consumerreports.org/cars/car-recalls-defects/nht...
That's a fair contention: I've never personally driven a vehicle with such mechanism (perhaps owing to my economic inability to own such a device)
I appreciate your informed opinion!
> but that means little compared to an individual who already drives safely without it.
https://udayton.edu/blogs/erma/2022/06/lakewobegon.php
My 'boomer' relatives refuse to wear a seatbelt, regardless of how many fines they've incurred, exactly because they already drive safely.
Every possible change to a car’s safety system will cause some deaths and save others… look at airbags.
The aggregate stats are the most important things to look at, especially for something like AEB which has the ability to save pedestrians, which as of late are getting mowed down more and more frequently.
Agreed that overall it can be an improvement. But I've clocked up a good 40K km since I got rid of the car that was equipped with defunct EAB and I'm fairly sure (but of course can not prove it) that I would have been in an accident by now on account of it triggering falsely. Conversely, I've never had a problem with airbags deploying without cause and my seatbelt is on just to move the car 10 meters because it is deeply ingrained and I refuse to drive - or be in - a car without it.
But faulty EAB caused more trouble for me than any other safety feature to date and I really wonder what caused Mercedes Benz to release this feature in such a horrific functional state. It is beyond belief how broken it is and I would love to see their internal stats for how often their EAB caused - not prevented - single vehicle accidents.
There needs to be a tiered license then - if you pass the more rigorous exam you can drive cars without the extra "safety" features. Making AEB mandatory would increase the survival rate of poor drivers and punish people who actually pay attention on the road.
> EAB will cause some accidents and prevents a whole lot more, and that's enough for the insurers.
It should be enough for everybody on the road. People always say "don't be a statistic" but we are all part of these statistics whether we like it or not. AEB can also protect occupants of other vehicles, so using it or not is not a choice whose consequences only affect the individual making it.
Wasn't AIG an insurance company? They don't seem to be good at math, much less statistics.
Exceptions, rules and all that.
Which is funny considering the explanation that was presented to me about why is it legal in the US to ride a motorcycle without a crash helmet. The explanation is: in case of an accident, the motorcycle driver has a lower chance to survive without the helmet. From the insurer’s perspective paying out a one time death insurance might be cheaper than paying someone money every month for the rest of their life.
No idea how much truth is in that, though.
> why is it legal in the US to ride a motorcycle without a crash helmet.
This depends on the state:
https://www.cdc.gov/transportationsafety/calculator/factshee...
> As of 2021, 18 states and the District of Columbia had universal motorcycle helmet laws, 29 states had laws requiring helmets for certain riders, and three states did not have any motorcycle helmet laws.
Interesting! My recollection was that it was a non-explosive tensioner, but maybe I'm wrong. Thinking about explosive tensioners reminds me of an EOD tool I recently learned about – the rocket wrench. https://www.youtube.com/watch?v=11qa5LFB6zI [1:36]
There are spring loaded ones too but they're not as powerful or as fast.
https://laughingsquid.com/a-slow-motion-video-of-a-pyrotechn...
Was that a VW? Sounds like their PreCrash system. When it detects impeding collision, it tensions seatbelts and closes windows.
So do you have to replace the explosive even if you don't end up crashing?
If it's an explosive one, yeah.
I believe that the statistics are that most people don't brake anywhere near hard enough in accidents, because you're just not used to pushing the pedal that hard.
Yes this is what we see in the car control clinics.
I've been involved in these since the 90s. Back then the braking exercises were about teaching threshhold braking since most cars didn't have ABS.
I speculated that once ABS becomes mainstream we'll need to remove these exercises from the classes since there won't be anything left to teach, just press the brake and the car will do all the work.
I never foresaw that "just press the brake" was going to be so difficult for so many people that the braking exercises are still to this day very useful and most people fail badly the first few attempts!
For anyone reading, if you've never braked hard, braking hard in an ABS car means push on that brake pedal with the full force of your leg muscles as hard as you possibly can and then keep that pressure on, don't let go.
(Also, please sign up for a car control clinic.)
In the UK the driving test includes an emergency stop which means everyone has to learn to do one. It definitely is surprising the first time you do one. It's worrying that there are drivers out there who haven't done one and don't even think they are capable of doing it when necessary.
Yes, unfortunately in the US getting a driving license does not include any car control training. It is really criminal negligence by the part of the government/DMV.
This is just one of those suspicions I have had for many years and done nothing to verify, but I think part of it is probably that everyone sits too far back from the pedals.
I have my seat adjusted so that with my back and butt against the back of my seat (so I’m braced against something) I can push the brake pedal and have it hit the bottom of travel just a bit before my knee locks without having to take my heel off the floor and try and press it with my toes or something.
My wife is almost a half foot shorter than me and when I drive her car I don’t need to adjust the seat at all. I find a similar situation driving most people’s cars.
So, y’know, I think people literally can’t apply the brakes all the way… because they can’t reach the bottom of the pedal travel.
I suspect it's more like people brake at first like a startle reflex, but they only brake really hard once their brain does the calculus and full panic sets in. Then they can bend the brake pedal.
On a motorcycle this is actually a lifesaver that's a very natural reflex because you want to load the front suspension and tire before you almost lock it up.
I'm not sure what the exact amount of force is, but I took a more performance oriented driving course that included some hard braking at highway speeds.
It's fairly violent. In normal driving we never do this, it would cause someone to rear end you and it's not necessary. So we have lots and lots and lots of practice braking at a moderate pace. Braking fast in normal day to day life is maybe a 70% force. We're used to trying to think how much pressure is the right amount of pressure.
What they told us is basically to try to break the pedal off. I think it's the easier way to remember, if the crap hits the fan, just put your foot through the floor and get stopped. The less thinking and modulating, the better.
To add on, most people have never experienced what “full braking” is in any car since it is so uncommon to even use half of the ultimate braking power a car can deliver in a non-track day or non-emergency situation. Even a modern Toyota Camry can sustain on average >1 g of deceleration force from 70 mph to 0.
Thus, people think they have applied full brakes but in reality they are only 60-70% of the way there.
Maybe people who learned to drive on pre-ABS cars?
I can remember clearly my thoughts during before bumping into someone in front of me: it was quick alternates between “BREAK NOW” and “beware slipping”. Traffic was quite slow but but I was terrified to not being able to choose what voice to listen in my head, it was a very strange dissonance experience.
I braked too softly and ended up in the bumper of the car in front of me. The rental car did had ABS and I knew it but didn’t thought of it during the breaking.
in panic mode brain doesnt work well
There's a similar one among sailors: "The mark of a great shiphandler is never getting into situations that require great shiphandling."
Anyone operating a machine in a way that doesn't leave much margin is, in fact, not skilled at operating that machine.
It's really interesting watching videos of emergency responders in my country (the UK) either showing how their advanced driving looks in practice, or in later stages of training when they're explaining their thoughts to an instructor as they work.
Their progress through traffic is markedly faster without greatly compromising safety. They aren't making violent manoeuvres, they accelerate pretty aggressively but other than speed limits they aren't breaking many rules that apply to everybody else. They are paying a lot more attention to the environment and applying excellent judgement from what they learned. For example rather than weaving through traffic like a Hollywood movie, they take clear bold lines and hold them, which gives them better vision of what's ahead and more options when things go wrong. It's doesn't look as flashy, but it's very effective.
What is the raw cost of AEB to the manufacturer? You need some sensors, a computer, and a way to actuate the brakes. Doesn’t seem like it would represent a significant fraction of the cost of even a cheap car.
They also already have that computer and most of them have the sensors too. The reason so many cars come with lots of these features is that the base electronic package that the maker uses just comes with it, so the R&D is already paid for and removing it would be more labor than the components cost. After all, it’s a $5 radar, a $3 chip, a $1 camera, and $0.02 of wiring.
Sadly, most anything is disorienting in crash situations. I recall how disoriented my mother was when a car came from the side and almost hit us. She tried braking and swerving, but that just caused a hydroplane and when the wheels got traction again, jerked us to the side. She was convinced the other car hit us. I was, too, all told. Until I saw that there was no impact on the side of our car.
Which is to say, you aren't wrong. I prefer folks knowing what their stuff can and will do. I am not clear on what that really says about deploying AEB. How many folks have no idea why their seatbelts lock up if you pull them all the way out, until you let them retract all the way? Very useful feature that they should not wait for everyone to know about before sending it out.
Well, if it's mandatory on new cars, then kind of everybody knows its there, and then there's the fun of not realizing it's not there on older cars. While it's optional, it might be nice if they had a notice on the dash that it's enabled.
On the seatbelt locking, I learned the why when installing car seats, which seems reasonable, because that's what it's for.
With all due respect, it sounds like your mother oversteered more than anything. AEB has little to do with that.
The propaganda that hyperfocuses on speed (ignoring all other factors in the system) has been very effective.
People seem oblivious to the reality that vain platitudes will never convince aggressive drivers to drive more safely in traffic.
Police are guaranteed to obsess over speed. It's just like the war on drugs: officers can measure and record criminal behavior! Easy money. Not once in my life have I even heard of the other law that is posted on signs - slower traffic keep right - being enforced.
Traffic flow is as important as any other variable in this system. Crash intensity is zero when there is no collision.
But odds of rollover and braking distance are proportional to your absolute speed. So is crash severity when you're impacting a tree, animal or pedestrian.
The problem is that there are more than two drivers involved:
There is you, and the slow driver blocking traffic in the passing lane next to you.
Then there are the 8 cars behind you, and at least one of those drivers is making everyone unsafe out of frustration.
You can't do anything to change that driver's behavior. The situation persists until the system is altered.
Typically you're not at high speeds when those risks are there. And when those risks are there on high-speed highways, they're typically due to poor road design (sightlines, etc.) and not the instincts of otherwise good drivers.
> I think this is human nature at play
Without a doubt it played a roll in suppressing what had happened. There is a weird dynamic where my wife won’t say things to upset her mom in particular ways, and this was one of those cases. They kind of mutually pretended the near-catastrophe away by behaving as though it was no different from a speed bump.
This is why I’m pretty sure we’re really not much different from a gorilla or something. We’ve got some sophisticated brain circuits but I have a deep seated sense that most of our brains are still the same old meatballs the other apes are packing around.
It was one of the scariest vehicle incidents I’ve been in (thankfully). She really shouldn’t be driving; she seems almost senile to me, but her kids and husband are determined to pretend nothing is going on. She lives full time in Mexico so there’s not a lot I can do other than be grateful for that buffer.
I once was in the car with an American friend who was looking sideways while talking to me and would then swerve around cars in front at the last moment. Very annoying and quite scary. I asked him to stop the car and got out. There is no way I'll trust my life to someone who isn't 100% on the ball when it comes to driving. Good for you you did not end up in an accident there, and it must really suck that you are essentially powerless to do something about it given the family relationship. But, at least one very positive story about automatic emergency braking. I think such features are quite possibly a net positive from a statistical perspective but for me personally they are a no-go.
They should be equipped with cameras at both the front and rear and ship off a recording of what was occurring on both cameras in the seconds leading up to and following every time the emergency brake systems engaged.
Only then can they determine what percentage of a specific emergency brake system's engagements actually make things more dangerous or are responsible for accidents.
Then they should hold the auto makers accountable, and require them to fix or replace fault vehicles
Sorry to hear about the accidents. One part makes me wondr (tangential to your stories). Has honking really ever prevented any accidents? I don't even know why a car needs a horn, 99% it's abused out of frustration, and I cannot Eben imagine what that magic 1% situation is that requires car to have a horn. Honest question.
I'd love to drive one, but they always seems like death traps to me haha. Makes an Elise look like an SUV!
I agree - the low-slung structure is excessively unfit in the era of SUVs and crossovers. I have dreamt of owning one for decades now but the price often gives me pause - I wave away the personal safety considerations by telling myself that it is probably safer than a motorbike :) .
I like to have fun while driving everyday, but yea you gotta be safe / courteous.
Among other things, data showed mandatory helmet laws may have had a perverse impact on motorcycle fatalities: http://john-adams.co.uk/wp-content/uploads/2006/motorcycle%2...
Haven't read it completely or paid attention to the data yet, but the conclusion got me hooked:
> Conclusion
> The foregoing review of the statistical case for helmet legislation suggests that the faith of the BMJ in the reliability of the statistical evidence on the effect of helmet legislation is misplaced. But debates about sophisticated statistics should not be allowed to drown out the voice of common sense. The risk compensation hypothesis is essentially common sense. There is a wealth of evidence from everyday experience that suggests that people's behaviour is influenced by their perception of risk. People tend to be more cautious when up high ladders than when up low ladders. They tend to take more care when standing on the edge of a high precipice than when standing on a low kerb. They tend to slow down when they encounter bends in the road or patches of fog and to speed up when the road becomes straight or the visibility good-and so on. The possible illustrations of the phenomenon are countless.
> There is an impressive amount of propaganda designed to encourage people to believe that they are very much less vulnerable when using seat belts and motorcycle helmets, and common sense suggests that driving behaviour will be influenced by safety devices that diminish the user's sense of vulnerability. What unaided common sense cannot predict is whether the behavioural changes induced by a safety device will partly, completely, or more than completely nullify the intended effect of the device.
> The statistics which the NHTSA considers the most compelling evidence in support of helmet legislation are displayed in figure 1. The state which Watson et al (1980) consider has had the greatest increase in deaths as result of helmet law repeal is Utah.
> When the distorting effects of the analytical methods employed by the NHTSA and Watson et al are removed, their data (with the caveats discussed above) suggest that the effect, if any, of helmet legislation on motorcycling fatalities is perverse.
I'll have to read the paper with more attention later to see if I agree with this, but it's a fascinating hypothesis regardless.Also, this is tangential, but I'm always impressed how different conclusions can be based on the same sets of data. Statistics is a truly fascinating field of mathematics because it's still an objective topic (regarding numbers, in isolation) but becomes very human and subjective when it comes to the different conclusions, transformations, and visualizations for the same data sets.
> make cost/benefit tradeoffs that decrease the safety of those around them
And yet we see that truck and suv purchases are through the roof with buyers regularly expressing the sentiment that they would rather be in a large vehicle to keep themselves and the other occupants safe. Generally the other side, which is fuck the pedestrians, cyclists, or other drivers who cannot afford to drive a full sized truck to starbucks, is left unsaid.
> buyers regularly expressing the sentiment that they would rather be in a large vehicle to keep themselves and the other occupants safe
Yes. This is precisely why this sort of decision cannot be left to individual consumers -- the end result is an arms race of ever-larger vehicles and dramatic increases in fatalities for the rest of us.
The majority of people killed in crashes are impaired by alcohol or drugs or both. 1/6 of all fatalities are pedestrians, and another 1/6 are motorcycles, despite only being 1/10 of all registered vehicles.
There are tons of modes in this data. Assessing all crash fatalities as some sort of "general" problem you can target misses this fact.
The single greatest thing you could do to reduce fatalities is to eliminate drunk and impaired driving.
This comment is irrelevant to the discussion of the AEB, since, AEB doesn't prevent anyone else from working on eliminating drunk and impaired driving. If anything, AEB would help mitigate some of the impact from the drunk and impaired driving.
Agreed that eliminating drunk and impaired driving is important. (For example, if there was a well-tested and reliable technology that could reduce drunk and impaired driving, I would be in favor of mandating it in new cars.)
But this is not mutually exclusive. For one thing, we can do two things at once. For another, automobile fatalities typically have multiple contributing causes. If we had stronger safety regulations around vehicle height, stopping distance, windshield visibility, etc then some number of the crashes involving alcohol or drug use might have been injuries or close calls instead of fatalities.
> it makes perfect sense to tightly regulate automobiles
Not the way we do it, it doesn't. What it does make sense to do is put the liability for harm where it belongs: on those who actually cause harm. But the vast majority of our automobile regulations--I include traffic laws in this category--do not do that. The regulation proposed here is an example.
Nonsense. Better than worrying about liability of the harm that happens, is reducing the amount of harm that happens. To argue otherwise is silly.
All of the safety devices you mention were bought by many consumers before they were required by regulations. Why? Because they are obvious benefits with no real downside. In fact, the only reason regulators were able to require things like anti-lock brakes and seat belts on all cars was that the vast majority of cars already had them, because the vast majority of drivers already recognized their benefits. They didn't need regulators to tell them.
The systems under discussion here are different, because they have the capability to do something that none of the other safety systems you mention can: take control of your car from you. Anti-lock brakes only work when you press the brake pedal. Seat belts and rear view cameras are passive. Automatic systems that can press the brakes without warning and without you expecting it are very different; what if the system makes a mistake? Multiple other posters in this thread have described such mistakes.
These systems are more like rudimentary self-driving cars and should be judged accordingly. They are not just passive safety features with no downside.
The car industry fought seatbelts and airbags.
https://www.nytimes.com/1975/04/06/archives/industry-resists...
Your claims about widespread voluntary adoption appear false.
https://www.history.com/news/seat-belt-laws-resistance
> At the time, only 14 percent of Americans regularly wore seat belts, even though the federal government required lap and shoulder belts in all new cars starting in 1968.
As a sibling post said, this is not that accurate, all told. I specifically remember similar argument against anti-lock brakes when those were first made.
To your credit, I have never heard anyone say bad things about rear view cameras. It is scary how many lives that has to have saved by now.
Now, I have heard complaints on cruise control. I don't see that going away any time soon.
Yes, they do. Some even activate hazard lights when coming to a rapid stop.
That’s what a disabling trigger on the steering wheel would be for, one of the reasons at least. There may be situations where I want to ensure it is off, such as in an environment where it has often detected false positives in the past.
Seconded, better yet a public registry of locations and situations where these systems go off so that manufacturers can include these in their tests.
Usually you can disable it but on a per-ride basis and it takes a lot of menu scrolling and tapping to do it again. It defaults to 'on' when you start the car, and there is no physical switch to disable it permanently.
Yeah, I meant it's weird you cannot just disable it and have that saved across starts. Mine allows me to disable those features and they don't turn back on when you restart the vehicle. I don't think AEB is a feature that can be disabled though. Just the collision avoidance feature. They are similar, but not the same. AEB is supposed to only trigger when a crash is imminent while collision avoidance is what triggers when it thinks you are about to have a crash that can be averted. I've never had AEB trigger, but CA triggers regularly enough that I know when to expect it now.
I imagine there is no permanent disable because it is a "safety" feature. But at least on my Subaru disabling it is a physical button. No scrolling and tapping required. On the other hand, it also works well enough that I haven't felt a need to disable it.
I too have a Subaru, my emergency brake has only engaged when trying to parallel park a bit too aggressively. I must be a very good driver! But perhaps it has saved my bumper a few dents
so if someone merges over into your lane where you are it doesn't activate? I must not understand the point of this feature then.
On my Honda if a car merges in front of me and is accelerating, it maintains its speed presumably on the assumption that the separation will continue to increase. If a car merges in front and is slower than me, i.e. separation is decreasing, it will brake in an attempt to maintain the set following distance.
In practice it does leave a bit more following distance than most human drivers, but your worry of constantly getting cut off until you come to a stop in the road is not a real thing. This is partly because the car modifies the set following distance based on speed, so that the same setting is much closer at e.g. 5 mph than at 60.
Someone else explained, but AEB is not cruise control. It tries to stop you if and only if it genuinely thinks you are about to crash into a stationary object or and object traveling at such a different speed that it might as well be stationary. Ie, a potential crash, not an everyday lane merge situation
It’s for when your mother in law hurtles your family towards a truck in front of you with no indication of trying to slow down (personal experience). The car does its best to stop acceleration while maximizing braking force safely.
I've had a car that apparently has "automatically breaks" because the darn thing spent more time in the shop than on the road.
What you get for buying a bombcar.
My mother reports that feature as really bothersome in a new Land Rover. So much that she just doesn't engage the cruise control.
Yes, because the automatic braking 'feature' can cause you to crash.
Keep in mind that Not every car brand has the executional prowess of Tesla. As the others have mentioned I’ve never experienced phantom braking on my Subaru while driving.
It hasn’t even done the object detected warning beeping when driving on actual roads, which I assume happens prior to the braking. The times I’ve had that warning are parking head on against walls and approach fast, and on one particular 15 mph off ramp that curved sharply such that one drove directly at a sign before turning sharply. It was understandable that it would alert there. No braking.
Assisted Lane keeping is another story, not a fan. Does not account for a ton of things such as the car or truck in the next lane is almost over the line.
More precisely, cause others to crash into you.
Automaker implementation and the chosen hardware to do the implementation on. Meaning that, for example I drive a Honda which has a camera and a radar. The camera is for lane assist from what I gathered and the radar for ACC and AEB.
If I accelerate into traffic and get close to a car while accelerating, it will start flashing a warning light and if I still don't stop accelerating or braking, it will intervene, but if the other car accelerates it doesn't do anything. So the radar setup is quite well done and it gives very few false positives.
The lane assist system is a bit bad though because once it identifies the lane markings it does not take into account potential hazards that want me to change a lane or exit it for a few seconds to avoid road hazard lights or something similar.
The point I am trying to make is that there should be a high standard to what drives the AEB and it should not rely on cameras, which seems to be what automakers want to implement driving assisted features to save on more advanced tech. And somehow I don't think they're gonna pony up for high-speed cameras that can replace a radar at accuracy.
Maybe the intention is good, but most likely the implementation will suck.
Alternatively, it should have a wider field of view and if nothing is obviously very wrong let the driver move the wheel around without intervention.
> applies too much resistance when attempting e.g overtaking
It doesn't turn off when you indicate to change lanes? It does on the Hyundai.
Generally it does at least in my US model. "The system is set to avoid unnecessary alarms if the system detects the vehicle behavior is intentional judging from the operation of the accelerator or turn signal." from https://www.mazda.com/en/archives/safety2/active_safety/las/
There are still scenarios where it can be an issue, for example when you need to "peek" through the lane to see if an overtake is possible safely. In this case it requires driver judgement on when to activate the turn signal. You may activate early or when committed, I don't think either is necessarily wrong.
I don't know the other commenter's scenario but don't doubt what they described. To pull another example out of my hat: bendy mountain roads (e.g. western Connecticut, minimal elevation changes), car ahead is moving slowly due to lack of power, heavy load, etc. Assume driving conditions and road lines are clear, no camera or sensor obstructions. The system is automatically active while following the bend and will attempt to keep on course. As you approach a rather wide curve (lanes are also widened, with extra shoulder) it is an opportunity to ride closer to the centerline in order to see past the vehicle and overtake. If you do that without putting on your blinker first, the system will attempt to correct while you're in the turn. Depending on how you react this can be jarring as the wheel is vibrating and you have to decide whether to commit to let the system correct, correct the correction, or commit to the overtake.
I chalk this up more to my own lack of familiarity with the system early on and not a defect per se. The wheel vibration is a very odd sensation to me on turns likely because on older cars I'd start to worry about tires or mechanical issues.
Mine does too (Volkswagen) but it's been my observation that most people don't indicate to change lanes.
I have a Mazda and don't have any of the issues the commenter has including any type of resistance when changing lanes.
Definitely. As a New Yorker now living in the Deep South, it’s night and day. All of those features are off (including traction control) in the northern cities. You need to be on your toes in nyc especially.
But in the south? Not a single one of the original poster’s complaints are valid here with the exception of cars who cut you off and then speed away (no need to slam on the brakes, cruise control). Personally, the early brake warning has saved someone else’s property at lease twice.
I'm familiar with NYC and North to South Carolina traffic. The only time I have had issues outside of the city was on a drive up to Montreal, where the lane assist misread the lane shift on potentially icy road. Wasn't happy about that one, but it was easy to overcome. I've never had an automatic braking incident outside of NYC that I can recall.
Hyundai has been having engine failures, would not recommend as experienced spontaneous engine failure on a less than year old car. Check consumer reports or source below.
“In 2020, NHTSA announced civil penalties for the companies—totaling $210 million—for not recalling vehicles in a timely manner. In 2021, a whistleblower who provided NHTSA with information that led to those penalties was awarded $24 million, the agency’s first-ever whistleblower award.”
https://www.consumerreports.org/cars/car-recalls-defects/why...
The Ioniq 5 is an EV, can't imagine the engine failures on their ICE cars will have any effects.
If you use single pedal mode watch out for your lack of brake lights
Hyundai had terrible engine and transmission problems in the 2010s. A local dealer was offering 20 year warranties to get them off the lot. There was a Reddit post of a 300k+ Kia that had gone through something like 7 engines and 4 transmissions. They basically 4 new cars worth of engine and transmission parts.
They have great styling that’s comparable to German cars. I wish the US automakers could take some hints. From what I’ve heard the Genesis lineup is very luxurious.
Comma 3 is s the best, to be honest. Due to CAN-FD on Ioniq 5 the setup is a little bit more complex, but this is the best experience, though they are still working on their own AEB implementation.
I'm not sure how exactly it works, but for my Hyundai at least it doesn't seem to be "you will hit this if you do nothing", the alerts generally pop up if I'm actively accelerating. Maybe if I got closer it'd kick in, I've just never heard the alarms when I'm already slowing down (but should be doing so faster).
I know mine is camera-based. The car does not have adaptive cruise control, so no radar, just front parking sensors - but those are probably just used for the "the car ahead of you has started moving!" warning, which is the exact opposite of AEB.
It came from cheap unreliable cars they made in the early 90's. Those things were through away vehicles. By the early 2000's they were just cheaply made(material wise) but reliability has improved vastly.
is there a similar service in NL or Europe? NTSHA is really an institution that's lacking in the EU, the recalls and safety bulletin are a feature I would like to see in the EU.
The AEB in my 2015 Volvo applies some extra force to the brakes which exceeds just pushing the brake pedal all the way down. Also, according to Volvo statistics it seems that half of all drivers who hit another vehicle from behind do not brake at all prior to the collision. So no, they won't achieve it the old fashioned way all the time.
Ok, looks like a different sensor then, the MB unit uses Radar and optical.
Lane-keeping should only be active with cruise control active. You shouldn't be using it in city traffic, ever.
Automatic lane-keeping assist is generally turned on at all times with modern cars.
My 2017 Pacifica (built in 2016) has lane keeping on by default without cruise control. I'm pretty happy with its light touch, so I wouldn't expect it to really move me out of the de facto lane into the lane it senses, but it could try. I'll manually turn it off for snowy roads, but otherwise let it do it's job of supervising in case the human screws up. Otoh, other vehicles may have a stronger correction, or my least favorite where they just beep a lot.
It is on by default in my 2020 Mazda3. After giving it (and all the other safety features) a chance for several months it's muscle memory to flip off all safety features during city driving. Thankfully there's a single button that does that, just left of the steering wheel.
> The scenarios and questions you ask are easy to explore in the data.
This sounds like the beginnings of the McNamara fallacy: that there is some abstract "data" where all the relevant variables can be explored to decide the best course of action. The reason it's a fallacy is that data is itself only a reflection of conscientious decisions about what to measure and how to measure, and these are invariably limited to what we already appreciate and understand, not accounting for the things we don't know how to measure and don't know we should be measuring.
To be more concrete, let's say we follow your suggestion and focus on crashes. Maybe it's found that automatic breaking reduces the number of crashes so it's mandated. But maybe automatic breaking also increases the likelihood that a crash is deadly in some sub-population like pedestrians, construction workers, and motorcyclists. So overall we see crashes decline because a large number of minor crashes are averted but we also see deadly crashes increase because pedestrians, construction workers, and motorcyclists are killed more frequently.
So data isn't a free lunch. Every decision about what to measure and what to optimize involves some prior contextual understanding. A misunderstanding of the context means the data will misrepresent some aspect of the problem and you'll overoptimize yourself into a local maximum.
Again, you are contriving bullshit to scare folks. And you didn't look at the studies. 38% decrease in accidents along with decrease in fatalities. In the data. For the worries you are sprouting, the data would have to look very different.
I disagree. I've been a licensed driver for over 40 years. I had one minor fender-bender at 16 years old, learned my lesson, and haven't had another accident since. I'm an exceptionally good driver as evidenced by this, plus the fact that I've taken performance driving classes, simulators, and so forth.
I, personally, am a much lower risk than the typical driver. Putting a drivers' aid device in my car may well increase the danger to me and those around me, even while having it in other cars decreases danger.
BTW - my car's traction control is evil. It's a rear wheel drive car, without limited slip. So when I pull out onto the road from a driveway, if there's the slightest bit of sand on the road, I get enough spin to engage the traction control. The car deals with this by vastly cutting engine power. In Texas, where the regular surface road I'm pulling out to is commonly 60mph, and often faster, I need to accelerate quickly to avoid the guy approaching - yet the traction control has just left me wallowing, dead in the water, unable to accelerate strongly. Except in wet conditions, disabling the traction control is always the first thing I do when I sit in the car.
Every person on the planet believes they’re an above average driver. No one thinks they need these types of safety features in their car because they’re a safe driver and the problem is every other driver.
It terrifies me that people are so oblivious to their own flaws that they would willingly disable features such as AEB if they could. I’m extremely thankful for regulations that prevent such life-saving safety features from being disabled.
I thought I was a low risk driver, then when doing a lot of hours riding a motorcycle it became very apparent how incredibly terrible of a car driver I was. So much general threat and risk was invisible as a car driver contained in a big metal box.
Even if you would rather not have the feature in your car, wouldn't you rather the idiot about to hit you had it in his?
Unless you are proposing that they should require it in everyone's cars except yorus.
> If the number of crashes goes down, contrived scenarios don't really impact things.
Yes they do impact things.
Consider a pill that, given to a wide population of test subjects, reduces heart attacks by 20%. That's great! But when you look at the data more closely, it strongly reduces the risk for people older than 40, and it increases the risk for people under 40. Should you prescribe that pill to everyone at risk of heart attack? Of course not.
Does AEB have a pattern like this? Who knows! Someone should test or cite a good test!
TFA links a meta analysis. With a 38% impact, that is going to be tough to argue against. So... Consider it cited, but you aren't looking?
The question is whether anyone is actually interrogating the data in that way, not whether it can be interrogated. Large data-driven institutions are notorious for taking a narrow view of things.
The problem is precisely a lack of imagination, or perhaps more accurately, critical thought.
The parent is precisely engaged in the kind of interrogation that is necessary. If you are aware or an analysis that bears on his questions, now would be the time to share it.
This feels like sophistry to hinder progress. The OP is literally about studies done. Did you find problems with them? Or are you only imagining problems there could be?
The number of crashes will go up because of this "safety" feature. And then we'll be stuck with this "safety" feature for decades till the govt gets their act together to undo what they did.
I'll keep my 30yo car with no computers as long as I can thank you very much.
The confusion here is that some people realize they aren't "average". That is, while some policy or another is good on the whole, it can also be bad for them personally. For them, the chance of a wreck can actually increase. You're demanding that they sacrifice themselves to save imbeciles who probably shouldn't be allowed to drive anyway.
For such people, these policies suck. And it's strange that there are those that demand they either be happy about it, or at least be quiet about it.
Anyone that frames this as a sacrifice is an entitled person. Same as folks that argued against seat belts.
I think you missed their point. It's not that most situations are improved and some get worse, it's that those situations might be unevenly distributed among drivers.
A seat belt is basically the same for everyone. It makes things safer when you look at all drivers, and it makes things safer for just about any subgroup of drivers.
But a buggy AEB might not fit that pattern. Imagine a system that reduces crashes by half, but then adds an extra X crashes per million miles. If someone's unassisted crash rate is 5X, then this system reduces it to 3.5X and is a great help. If someone's unassisted crash rate is .5X, then this system increases their crash rate to 1.25X and is a terrible trade.
Your claim seems to be the group who are not helped are "good drivers"? ... except that doesn't feel like a well-defined group to me, and also is a group that almost everyone would say that they are in.
> it's that those situations might be unevenly distributed among drivers.
I think that may be true in all cases - for example seat belts and airbags requirements may affect people differently depending on height and weight?
They way you crash in terms of whether your seatbelt helps or harms doesn't depend on your life choices. Features likely benefiting drunk and distracted drivers and likely hurting aware ones are different.
What's your point?
On Wednesday, the National Highway Traffic Safety Administration issued a notice of proposed rulemaking that would see automatic emergency braking become a standard feature on all new light passenger vehicles.
mhb said:
And the actual numbers should include the opportunity cost (in lives saved or QALYs, if you like) of the increased price of vehicles with these systems. Could that money have been spent on smoke detectors or healthier food or ?
In response I said:
National Highway Traffic Safety Administration
Then mhb said:
What's your point?
Can you explain what my point was to mhb?
ChatGPT:
It seems like you're suggesting that the responsibility and scope of the National Highway Traffic Safety Administration (NHTSA) is primarily focused on improving safety on the roads, specifically through regulations related to vehicles, rather than other broader public health initiatives like smoke detectors or promoting healthier food.
Your comment may imply that while considering the opportunity cost is a valid point for an overall public health or policy perspective, it may not be directly relevant to the mission and function of the NHTSA. The NHTSA's mandate is to ensure the safety of vehicles and roads, so they would prioritize improvements and regulations in that area, even if those funds might have a higher impact if spent elsewhere from a wider public health perspective.
In other words, the comment might suggest that each organization has its own scope and priorities, and while we can consider overall opportunity costs, we also need to consider the specific roles and responsibilities of different organizations. However, without more context, this is just one possible interpretation.
Removing distractions like poor user interfaces would probably overlap with the improvements from emergency braking. So if you want to be throughout you would have to check if they still make sense with the other changes in place.
That's what I was saying, yes.
It's fair that you can never wipe out all variables. However, a lot of people (especially taeric) were dismissing any attempt to discuss if other aspects were important as "the NHTSA has data and you only have conjecture."
While the costs of phantom breaking are decreased if everyone has AEB, I don't think this makes the roads safer for me unless I want to buy a newer vehicle and I don't want to have to do that.
Feel free to look at the numbers in the PRIA [1]. Those 360 deaths are a significant reduction in fatalities from baseline in the types of collisions they expect to be impacted by this rulemaking. That baseline itself assumes some 65% of new vehicles will come with compliant AEB systems regardless of this rulemaking.
[1] https://www.regulations.gov/document/NHTSA-2023-0021-0004
>>> math.sqrt(42939.0)
207.2172772719495
Not quite pure statistical noise but close.
I literally do not care about 360 deaths. Now if we get to 361 deaths, I am concerned and will raise my pitchfork.
Another interesting question is how many of the other 24,000 crashes were serious. And how many additional crashes wouldn’t have been avoided but would have had a lot of kinetic energy removed before impact.
This is a pretty big handwave:
> The number of non-collision severe injuries (are we theorizing folks get their head banged against something when it brakes heavily?) is likely to be pretty small
(People can be injured directly by the seatbelt, and there's also whiplash to consider.)
> People can be injured directly by the seatbelt, and there's also whiplash to consider.
I don't doubt it's possible, just like it's possible for a seatbelt to trap you in a burning car... but common enough to offset a 38% drop in crashes? I severely doubt it.
If someone can demonstrate that people driving these cars have a dramatically increased incidence of whiplash or seatbelt injuries, by all means, but I'm comfortable saying the burden of proof is on them there.
It's fair to say that some percentage of cases could be avoided by driving more defensively but I doubt it is significant. There are plenty of scenarios that the systems simply can't handle gracefully such as narrow or shifting lanes (my experience). The safest solution might be to avoid such roads entirely but then I'd never be able to leave my borough.
Road throughput is inversely proportional to following distance, so if everyone doubles their following distance to account for / avoid spurious automatic emergency braking, then we’d need to double the number of lanes to keep road capacity constant, which would lead to more lane changes, and therefore accidents.
If, instead, we kept the roads the same, then congestion would increase, and with it, accidents.
There is a reason that random unexpected emergency braking is a favorite tactic for people attempting to force accidents in order to commit insurance fraud: It’s too expensive for everyone to change their habits in order to guard against it.
If we care about reducing following distance, we should be lowering speed limits. A car doing 60 needs a much longer safer braking distance than one doing 45.
Also, if you are following at a distance where an AEB activation will cause you to rearend the vehicle in front of you, you are too close to safely follow any vehicle, regardless of whether it has an AEB.
> from the perspective of the manufacturer they would probably say such mistakes should not by themselves cause problems unless someone else is breaking the law
The kinds of mistakes described elsewhere in this thread have not been of the form "system mistakenly brakes my car and someone behind me rear-ends me because they were following too close". So your comments, while valid for that particular scenario, do not appear to be relevant to what other posters are concerned about.
It's not only rear ending. The main road near my house has a sharp left turn with a sign on the side of the road. I occasionally get a BRAKE! warning there, probably because AEB mistakes that sign for a car in front of me. On a wet road, stepping on the brakes when the car is turning can easily result in understeer, which is particularly dangerous in that spot because of the large boulders nearby.
Yeah I was just adding to the things to consider. In my case it has never really confused inanimate objects, the worst was a curved road and a car coming "right at me" which was just at an angle. But it was a temporary slowdown due to cruise control, nothing major.
Car reviews are largely useless fluff - a real analysis of the differences between brands for this kind of thing could actually be useful.
That matches one of my two faulty EAB experiences nearly perfect.
I don't care if the law says the person who rear-ends me is technically at fault; I don't want to get rear-ended in the first place. If I can't control the driving of the car behind me, at least I'd like to be able to control my car to prevent the possibility of such a situation.
Redundancy doesn't work for design errors. You would need two diverse systems that have to agree before applying the brakes. This would also increase the "false negative" rate. It's a balance.
Why do you think these are design errors?
I have the same car and my experience matches yours. Lane assist is just a tiny nudge, and I've never had the emergency braking kick in. I wonder what causes the different behavior?
To be fair I was going 70mph at the time. It could be a small nudge but at 70mph feels worse. I also had only owned the crosstrek for about 5 minutes at that point and I couldn’t figure out how to turn it off as well, didn’t realize buttons were placed on the ceiling.
It was actually my first experience with the car, it happened on my way from the dealership to the office. It also didn’t help that I owneed the vehicle for 5 minutes at that point and could not find the button to disable it. I have since gotten rid of the car.
Yeah its part of my pre-drive procedure since it happened. I’ve never had any of my vehicles auto brake in a way that was helpful.
That's becoming increasingly harder as EuroNCAP mandates more and more of these as standard technologies that must be included. Manufacturers would much rather build one car and modify it as little as possible for all intended markets. So they design them with American and Chinese expectations for size, appearance, and features, and with European expectations for safety and emissions systems. We already can't buy new cars without rear cameras or forward emergency warning for that very reason.
I got what was available at the time. The golf r is low production car finding one to buy was hard enough.
The crosstrek the only orange one available had eyesight.
The atlas was used so there was only 1 option locally
The impreza its standard on anything over base, and the base was missing other things i wanted.
Ok, that makes sense. I just bought a 1997 car and had it rebuilt, about 30% of the price of a new one and none of the crap, roughly just as safe as the current ones and much less likely to be stolen. Not the most economical decision because of course it is still a 1997 car but it will hold its value well enough that I'm not worried about it and I'm not really planning on selling it anyway. But if I ever do crash it then I'll probably lose a bunch of money because the present day value is less than what I put into it. Otoh just the write-off on a new one would have been roughly the total invested.
It's very frustrating how tech can be mis-applied. But I also see the regulators point of applied statistics and that even if these systems cause the occasional crash they prevent a much larger number from ever happening.
I know what it was, but I won’t participate in naming and shaming.
The dealership tried to fix it several times, and I don’t know if this is a systemic problem, or a single vehicle issue.
Either way, I don’t trust the tech. It can be developed and tested against plenty of different scenarios, but we live in a very uncontrolled and fluctuating environment.
> People should absolutely be able to turn this thing off if they don't trust it.
Absolutely not. There are other people on the road. Families. You don’t get to kill other people in a preventable accident because you believe you’re an above-average human who can outperform a computer.
For what it's worth, I think you're misunderstanding the sentiment here.
The argument isn't that a human is better at avoiding accidents than an automated system, the argument is that, as an average human, I'm good enough and the marginal difference isn't worth the complexity and risk that the system introduces.
I literally couldn't care less how well the computer performs. I'm the human, it's my vehicle, I'm driving and I should have total command over the machine. I refuse to be dehumanized into some well-controlled variable in the name of so-called safety.
Think of the children?
Computer errors have killed entire families.
ABS improves your control in normal circumstances. There are (uncommon) circumstances where it will actually harm your ability to handle the vehicle, same as other mandatory systems like ECS and optional safety features like traction control. Some vehicles have ways to disable some or all of these safety features for control purposes.
The proposed rules include false activation tests, monitoring/reporting of unnecessary braking, and NHTSA retains the ability to order recalls if false activations are unreasonably common. It also includes a request for comment on whether manual disablement should be allowed at high speeds. Feel free to submit one if you feel strongly about it.
Can you elaborate on that? Everything I've read about the ABS technology suggests it always improves steering and almost always reduces braking distance. Exceptions include surfaces like snow and gravel where it actually increases braking distances. If there's more to it than this, I could change my opinion on the matter: might be wise to have the option to disable in some situations.
Which is why it should be monitored to ensure that the net result of the rollout is a reduction in accidents/fatalities/injuries/etc.
Your comment is very hard to disagree. But how are you going to test if there might be some interference of two changes? For example, negative effect from screen/touchscreen and positive from auto-breaking - isn't your next proposal to gather more information including video of driver's moves? I have a strong feel that an argument that the drivers must be video-streaming while moving will be equally hard to disagree.
Right, it's a slippery slope. Today they're making your car save lives, tomorrow they'll be draining your bank accounts and seducing your significant others.
The examples I cited were not hypothetical. The "tomorrow" you dismiss as some unrealistic joke is already here. Search HN for the news if you don't believe it. Want another? Not too long ago some car maker's data leaked and we discovered they were recording people while they were driving. Got any other bright ideas you feel justified implementing in the name of "safety"?
They're "making your car save lives". How? By taking away your control. You're not in command of your vehicle anymore. Don't forget it.
How many lives makes it worth it? Just one? Should it make an appreciable difference? I am hesitant to hand the reigns of society to safety nazi bean counters who are more interested in appearing to "do sometihng" than actually improving our lives.
Beware false dichotomies
Joebob’s car insurance premiums and taxes cover others’ car crashes. He is statistically impacted even if he’s a recluse without a car at all.
(He might even be statistically more likely to run into the car he’s not used to seeing on the usually empty road via the stop sign he usually blows right through. Who knows?)
> (He might even be statistically more likely to run into the car he’s not used to seeing on the usually empty road via the stop sign he usually blows right through. Who knows?)
That's not how statistics work. Again, statistics mean little to the individual.
The car doing the T-boning can see forward, and even slowing down by 10 mph can halve the lethality of the resulting crash. Full success isn’t necessary for benefits.
Or sub in a high speed rear-ending scenario.
I’ve never had a bad accident in a car, but the closest I have ever come was in my friends new VW with lane assist on a dark country rode on a bend when the lane markers vanished due to wear.
I started to get lazy and distracted as the car was mostly driving, my friend was talking at me and the assist disengaged, next minute we were going pretty fast towards a barrier. We recovered but it was close.
I’d just prefer to drive honestly.
The article cites a 38% reduction in crashes. Random anecdotes to debunk that are not super compelling.
Some habitual drunk drivers never have an accident and may even claim it makes them a better driver, but that doesn’t make it the right choice on a population level.
This reminds me of the fact that many foreign subway systems have outer doors (so you can’t just walk onto the tracks) and inner doors. Most people won’t just walk onto the tracks in America but the outer doors address inattentive people, tired people, and people who want to do the wrong thing.
In a lot of places the outer gate is only chest high which isn’t insurmountable but is enough of a barrier to avert 99% of the bad cases. I think the analogy to automatic braking holds.
> If you are a safe driver, it's likely to be meaningless.
No. You are sharing the road with others, of variable skill, and things like deer and pedestrians who can dart into traffic. I’ve been crashed into from behind while stopped in traffic; no fuckup on my end, but still would have likely benefited from automated emergency braking.
Your car insurance also reflects others’ accident rates. You shoulder some of the medical and fire and police costs via taxes. Lots of statistical impact even in a perfect driver or someone who never drives at all.
Ok, I see where you're going with this, let's mandate for bad drivers; now everyone is happy.
Even a reduction in speed helps - 10 mph can make a significant difference in lethality - and sub in being rear-ended if you prefer.
The point is even a perfect driver may benefit from safety devices in the hands of other drivers they encounter.
This is a good point, thank you. I don't think anyone here would argue against this feature if it was 100% accurate. However, mandating something that has many reports of erroneous activation seems like overreach.
Those would be good stats to mandate sharing with the proposed rules. Would likely cause them to rework the system if they started losing sales over it.
At the very least it sounds like people will benefit from actual tests and requirements about how the braking system must act.
> and punish people who actually pay attention on the road.
It can benefit people who have reaction times slower than silicon.
It can also cause accidents, and some people go their whole lives without getting in an accident.
Well, forgive me then. That’s the discussion I had back in 2016, when visiting the US the last time. Good to see there’s progress.
Hm I think it was something else entirely. If I recall correctly, it felt like a motorized ratchet pulling me back in my seat.
Would love to know more
So for example https://www.youtube.com/watch?v=RmGDpUnlMas is people learning to drive an ambulance in an urban environment, so, they're under blue lights with a siren, they're authorised (while learning and after completing training) to exceed speed limits, and break various other rules, but they're being trained to do this safely (or at least, more safely than most people probably drive normally)
https://www.youtube.com/watch?v=wjkBchxT5kw is a solo, commentating live on his driving in a response car rather than an ambulance, again he's be authorised to break the rules.
They do have a dash notice on the brake engaging. And cars come with a giant book that almost certainly covers it.
And these same concerns apply to antilock brakes, no? Getting used to the way those feel took time for folks that were there before.
I didn't mean to implicate anything other than oversteer there. My point was that being in an accident is disorienting already, with no extra tech needed. Such that needing full teaching on it is an overly large burden.
Unfortunately, unless the frustrated driver is allowed to drive unimpeded at 100mph+, they will continue to be unsafe and make others unsafe. If there is enough traffic for you to be the slow car on the road, then that unsafe driver will be causing safety problems for all of you. Bottling them up in a line of cars going the same speed may be the safest course of action you can take for everyone.
That's very often not the case. Usually they just want to go 80 or 90, but are stuck going 63 because 2 drivers are aggressively passive.
Bottling up traffic is not safer! Every driver in a line who has to slam on their brakes, for whatever reason, subtracts the amount of available reaction time left to the driver behind them.
Traffic engineers the world over would disagree that decreasing space between cars makes anyone safer.
What makes more sense is leaning on the raison d'etre of multi-lane highways: slower traffic keep right. Designed systems work best when used as designed.
> essentially powerless to do something about it given the family relationship
Yeah, the fallout of telling a person they can’t drive is unreal. It took a couple years to go from incredulity towards me to finally accepting that I’m not joking/grumpy/changing my mind.
Some bad driving is so egregiously bad though that it can’t be chalked up as a lapse of judgement or something.
There is a strange sort of entitlement to recklessness like you described (why in the world would you take your eyes off the road to talk!?), and we seem to take for granted the fact that we’re still alive despite flying around in metal boxes within feet of each other. I think some especially bad drivers are severely influenced by survivorship bias. They’re not dead yet so they must be fine/good at driving and different from the people who’ve had bad accidents already. But they’re only protected by very bad odds.
I have a similar situation with an 87 year old family member that still drives and I've offered to be on call for chauffeur service or to pay for cabs or whatever it takes but they will keep doing it until they either die from other causes or get into an accident, which in my opinion is only a matter of time. So far I've managed to get them to rely on me for two really long rides and at least they no longer go on the highway. Highly frustrating because on the one hand I applaud the person for being that old and still self-reliant but on the other hand I am seriously concerned about other users of the road who are effectively in danger. And short of having them declared legally incompetent there isn't much that I can do beyond offer help.
I would challenge the common sense aspect of wearing a helmet makes you feel safer. Like if kevlar pants were mandatory on motorcycles would the same author claim it would increase accidents?
It narrows your visibility and the main feeling of danger (imo) was all the cars that act as if you aren't there (ideally because they didn't/can't see you).
> This comment is irrelevant to the discussion of the AEB,
This discussion is about _mandating_ AEB, not it's mere existence or implementation, and of the overall value of that proposition; which, I am suggesting is low relative to all other problems and efforts spent mandating this would be better spent elsewhere.
> If anything, AEB would help mitigate some of the impact from the drunk and impaired driving.
Are these systems tested for this use case? I'd like to see that. Otherwise, this is a guess, and a bad one.
Regulations that restrict people who have not caused actual harm do not, in general, reduce the amount of harm that happens. But they do work great as revenue sources for local governments (speeding laws are the most common example but not the only one), not to mention as excuses for manufacturers to charge higher prices and squeeze out more profit from consumers. They also reduce general respect for laws and regulations among the public: people start to view driving, for example, as a game to avoid being caught doing things that are technically illegal but which the vast majority of people do anyway (such as exceeding posted speed limits). Fostering that kind of mentality among the public arguably increases the amount of harm that happens.
Furthermore, government institutions, particularly in the post-COVID era, simply lack credibility when they make claims about some government mandate reducing harm. In this case, as I pointed out elsewhere in this thread, NHTSA's claims about harm reduction assume that these systems will always operate perfectly and will never cause any harm, only prevent it. Such an assumption is not credible.
In short, I do not share your faith in the ability of government regulators to actually reduce harm, on net. More generally, I do not share your faith in the ability of governments to actually act in the interest of all the people, instead of benefiting the governments themselves and particular special interests.
I disagree. Especially for institutions that have as much easy to look at data as this. Total wrecks is easy to track. And if it is going down, regulation to help is working.
Paranoia on overreach is fine. Good, even. That is best approached by auditing and running the stats. Constantly. But to claim that this is a meaningful restriction feels wrong on many levels.
That said, if you are going to assert some of these claims, on what evidence? I'm restricted such that I have to wear seatbelts. I can't buy a new car without rear view camera. I can't buy a new car without antilock brakes. Headlights. Brake lights. I could go for a long time on this list. Are you claiming none of those make driving safer?
> The car industry fought seatbelts and airbags.
But, as far as I can tell from the article, the industry is not fighting this regulation.
The parent post claimed they and the American public voluntarily opted for seat belts and airbags, which as far as I can tell was a falsehood. It is a response to that claim.
What percentage of Americans who did not wear seat belts suffered or caused harm because of it?
Enough? Consensus appears to be 300-400k people saved since 1975. Charting motor vehicle deaths against miles driven reveals a stark improvement in mortality as vehicle safety improved.
https://en.wikipedia.org/wiki/Transportation_safety_in_the_U...
Emergency services and social costs of crashes are immense.
AFAIK cruise control is not a regulatory requirement.
I can see how I muddied that. My point is I have heard complaints from cruise control accidents. If anything, I would expect regulation against cruise control.
> AEB is supposed to only trigger when a crash is imminent while collision avoidance is what triggers when it thinks you are about to have a crash that can be averted
If that had been the case I would have still driven that car... very frustrating because - touchscreen aside - it was a nice and comfy car and I had planned to make it the last car I ever bought.
On my 21 Subaru I disable lane assist unless I’m on the highway. It’s too strong on the wheel, and there are too many construction related lane issues near me that have almost caused accidents for me multiple times.
Do you not have strength settings on the lane keeping? My 2020 Kia Telluride lets me control how strong the lane keeping is with 3 levels.
Also, at least on mine, it doesn't attempt to implement the full follow distance right away. Mine can tell a vehicle is accelerating away from me and won't slow down to fit the spacing on the expectation that the spacing will soon be achieved by the other vehicle. If the vehicle in front of me is going slower than my set speed and not accelerating, then it slowly will lag back to get the set follow distance. You do have to worry about aggressive cars that cut you off in heavy traffic as the vehicle will just let them and lag back, but I can at least adjust the follow distance to partially mitigate this situation.
OK, it's good to know that it won't stop. But if it slows automatically it will still result in the same thing. All the other drivers just continue to cut you off indefinitely.
Actually, maybe I misunderstood you. I thought you were talking about adaptive cruise but you may have been talking about automatic emergency braking.
Interpreting your original comment in that context: AEB only kicks in at a point where maximum braking will result in you barely not hitting something in front of you, or maybe even hitting it at a low speed. Modern cars can stop very fast indeed with maximum braking so in practice this threshold is never crossed under normal driving conditions unless the software returns a false positive. It's not about maintaining a safe following distance—it's about stopping the car at the extreme edge of the safety envelope, beyond which no manual or automatic intervention could prevent a violent crash.
False positives do happen but I think they are very rare or practically nonexistent in good implementations. I've never had it happen with my Honda.
I didn't mention Tesla. FWIW mine was a Mercedes and it was the most dangerous car I've ever driven with distance. I've never had a car that actively tried to crash before.
My Kia likes to trigger the ~~AEB~~ Collision Avoidance warning when people are turning off the road in front of me. It's really only triggered when it _thinks_ a vehicle is in my path, but it's really exiting the path and just still in the periphery. I get about as many false positive type events with the lane keeping feature. I've learned, the hard way, that not all lane markings are to spec. The worst are when you have a side road and the solid white line curves off the road but they don't have a dotted white line continuing straight. The car loves to pull to the right in those cases. Not hard, but enough that you know it's being daft.
Yeah the lane keeping for sure is bad because even if perfect there is a lot of real word situations where one needs to deviate from the center of the lane.
So unless the vehicle is fully self driving to the level of an alert good driver, it’s going to mess up. And lane keeping is definitely not claiming to be that.
The lane departure warning beep is fine even although its 90% false alerts for me due to non-lane lines or snow etc, and 9% when two lanes merge and I get lazy about the turn signal with no cars around. But if I were to get more tired or something it could be a helpful alert of last resort to stop driving, hope I never wait that long though because that would be dangerous.
But I often see them used as an "apology blinker" (when they turn on the indicator after they've started to change lanes or turn).
Makes me laugh every time. I mean, at that point, why bother with the signal at all?
The engine failures are mostly for combustion, and have resulted in multiple deaths, including nearly my own.
Ah OK, in my car AEB and SCC ("Smart Cruise Control" as they call it) are the same sub-system - when one broke due to the no-input fail-safe, so did the other. The radar is behind the badge and works in tandem with the camera I believe.
By the way, no adaptive cruise control does not always mean that there is no radar. Some car still keep the radar and use it for AEB, but have ACC disabled I’d you do not purchase it when configuring the car.
Interesting; when did this happen? From my experience, ~2017-era cars had it only as part of cruise control.
Keeping it on at all times sounds like a terrible idea for cars used in the US, because it's a developing nation with extremely poor road quality. In developed nations, it's probably OK because you can count on the road markings being good almost everywhere, but not in the US.
I think it only turns on above certain speed and yes, it can get easily confused by various road imperfections.
But yeah, you need to make sure there is a physical button on the steering wheel or somehere on the dashboard to easily enable/disable it. Otherwise it can be annoying.
It’s not in either 2023 model year vehicle I own, but I have seen it on full time in another. But I have never seen it active below speeds of about 35mph (i.e. most city driving)
It's not "contriving bullshit" to give some examples of how a single average number could be misleading, clearly labeled as hypotheticals.
It is literally the definition of FUD here, though. These aren't things that have happened. And the data is ridiculously clear on progress. To paint these doubts is pretty much only to spread fear and uncertainty.
I think they have a great point though, accurate measurement of the types of incidents is just as important if not more than the total.
This conjures the fear that the agency in question is bad at this. Do you have evidence that they are? Can it happen? Yes. We could also have it so that no wrecks from minorities are counted, such that we are blind to how they are impacted. Would be terrible. No evidence that that has happened.
An average person has 3-4 car crashes in a lifetime. Having 1 minor accident in 40 years of driving is above average.
The only time I ever came close to having a crash is when lane assist tried to drive in the dividers on a highway with construction going on, so lane assist is the first to go in a new car.
The average person also has their wrecks at the beginning and end of their driving. Having 1 while in late 50s to 60s isn't that unlikely, either. And sets you on the average track.
> Every person on the planet believes they’re an above average driver
Not everyone. I don't think I'm an above average driver at all.
I do a lot of cycling and it's amazing how much more aware you are when any minor bump is going to cause you injury. I've avoided accidents multiple seconds in advance because something about the way someone was driving made me nervous, it's like you get a sixth sense.
visibility is also dramatically better on a bike
Riding a bike (motorcycle or bicycle) should be mandatory for learning to drive a car.
You're not listening. The 38% number can be true and there can be particular groups that are hurt more than helped. Nobody is trying to "argue against" the 38% number.
In a cost benefit, if the number drops 38%, you are going to have to work to show it isn't worth it. What would the number have to be?
Can they make the system even better by finding cases it didn't help? Almost certainly. Progress isn't a straight line, but increases are increases.
The latter, which is why the actual studies would be most welcome.
But more to the point, there is a track record of narrow interpretation of studies and missing of the big picture. It’s hardly sophistry, it’s critical thought.
Is there a track record for this specific organization? Or there are a lot of publicly known tech companies that have missed the big picture?
Insurance companies have a pretty good idea of who the "good" and "bad" drivers are. It's a question of statistics.
Well specifically the idea is that it's so uneven it might increase the danger for certain groups, and seatbelts are pretty good overall. But bringing up airbags is a great example. If you're short enough (or possibly if you're tall enough), a standard airbag is more dangerous than nothing at all.
Yeah. Duh. If you had included, in your prompt, the NHTSA quote that I quoted, ChatGPT might have been more helpful to you. Viz: On Wednesday, the National Highway Traffic Safety Administration issued a notice of proposed rulemaking that would see automatic emergency braking become a standard feature on all new light passenger vehicles.
That point being: maybe NHTSA's advocacy of a requirement for all new vehicles is beyond the scope of its mission and it should confine itself to providing information to the policy makers who might possibly understand the tradeoffs.
Depending on where you live, if you leave a big gap someone else will just fill that space and you're back to the "short following distance" mode again. Rinse and repeat...
And then when the car in front has to tap the brakes for whatever reason, the tailgater has to slam them to avoid a collision, the person tailgating them also slams theirs, and you get a wave of braking propagating backwards, eventually creating a traffic jam.
Human errors have killed many, many, many more.
It should be optional period. You want to save lives how about disable stolen cars, so joy ride Kias dont kill pregnant mothers. Imagine emergency braking a F-350 with a gooseneck and two tractors on it. Does not end well.
Also wet conditions, and worn concrete, etc. Worse, the control systems behind ABS systems are not equally effective at all speeds either, and typically have regimes where they introduce significant amounts of yaw oscillation during maneuvers. That's not good for the panicking driver and can result in them being unable to control the vehicle as effectively.
Note that these are all very niche concerns and I think ABS is a fantastic safety feature, but it's still a compromise to improve the averages. This is true of virtually all safety technology.
> might be wise to have the option to disable in some situations
Snow and gravel are the only valid situations where ABS can increase braking distance.
However, disabling ABS is still a really poor idea.
ABS increases braking distance in the snow because it allows the wheels to rotate so they can grip the road. ABS allowing your front wheels to rotate is why you have any marginal ability to steer and brake simultaneously. Without ABS in the snow your front wheels will lock while steering before your rears and you will lose control of the vehicle.
On top of that, disabling ABS usually disables the vehicle's entire suite of driver aids like traction control and stability control which are extremely helpful in snow.
I can't think of any situation, however contrived, where I would willingly disable ABS on a vehicle in snowy conditions... unless you were trying to slide a car around.
I agree with you. I can't think of any situation I'd want to disable those systems either.
Why should drivers be video streaming? I'm just suggesting that regulators should compare accident rates of cars with and without automated breaking. I'm not saying we should spy on anybody, surely regulators already get this data.
> Why should drivers be video streaming?
Because nobody has drawn a red line and Ron Jones in 1967 has shown us how vulnerable our freedom is.
We already has a lot of mandatory non-FOSS software in cars which use to phone home which effectively is a streaming (not a video yet, but amount of data is never decreasing). And a lot of so-called safety (surveillance in reality) devices are around us nowadays - phones exposing a lot of uncertain data about us, bank cards which allows banks know where we are right now, etc. If some crackpot politician proves to safety-demanding influencers that video streaming from every moving car helps us to prevent $bad_things_list so that kind of demand is totally possible, especially if using cars for mass killings would become more common than mass shootings.
> I'm just suggesting that regulators should compare accident rates of cars with and without automated breaking.
It's impossible because you have not 100% of the data.
“You are statistically more likely to die prematurely and painfully if you eat polonium” sure matters to individuals who eat polonium.
I don’t think you’re clear on the definition of “statistically”.
I don't think you are clear on the definition of statistically.
Citing things with >90% probability does not, actually, prove your weak point that "all statistical probabilities are individual certainties".
Little hard to respond to the snark but I’m just glad I was paying attention and not on the phone or else I might be dead.
Cheers.
No snark, and every avoided accident is a good thing. Policy makers at the national level just have to weigh all the cases over individual anecdotal ones.
This often means one person’s risk goes up while thousands of others’ goes down. Sometimes that’s hard to balance, but if the 38% reduction in crashes cited in the article is true, this particular decision seems fairly simple.
Good idea. People who think of themselves as good drivers are some of the worst usually, so we'll start with those.
Defective airbags blew shrapnel in folks’ faces, too. Anecdotes are of limited value here; I’m sure you can find the occasional person who has been trapped in a car by a seat belt and burned to death. Policy needs data, not unfalsifiable claims of being a perfect driver hobbled by computer interference.
The article notes:
> A metanalysis in 2015 found a 38 percent decrease in collisions for vehicles that utilized such a system, for instance.
Including those that drive cars with aeb.
I like the second one better. It's priceless to hear how slightly he accents unFORtunately he hasn't seen me after some driver blocks him for about 28 seconds despite blaring sirens.
Sorry to hear about that situation :( Here’s how it played out in my grandmothers life: she was fiercely independent until she (likely) caused a crash in her late 80s, ended up in hospital, then was never properly healthy again, she spiralled down gradually. It’s just not good :( hope that doesn’t upset you.
It doesn't upset me and that's exactly what I think may well happen because it has happened like that many times before with other people.
We're going through a similar situation with my grandpa. He's a great guy and has driven so much through his life, but he's almost 90 and his body just can't handle the reaction times anymore. Still, he insists on driving, and gets irrationally angry if you even come close to questioning his ability. Like, he is a sweet man and is normally very nice, but the second you mention his driving it's like a switch flips and he is suddenly a child with a tantrum.
It's gotten bad enough that we have to tactically lie to him to keep him from any unnecessary rides, but he will not stop if he doesn't have a wake-up call. Luckily he only drives very short distances which are all slow roads, but still, it's frightening to just be waiting for the day something bad happens.
I wish I knew how to bring the topic up in a way we can really talk about it. I have tried being extremely empathetic and non-pushy and careful, which has helped with other topics, but just not on this one. If anyone has tips or ideas I'd love to hear them!
My father is 87 and drives well. He is cognizant of his age and tries to only drive during the day and relatively short distances (I like 90min away and he is not visiting).
My fear is that if there is a decline that I won’t see it, and at 87 he isn’t interested in a new car with more safety features.
> and at 87 he isn’t interested in a new car with more safety features.
The safety features of a new car might not offset the safety of the muscle memory of driving the old car. At that age and with declining capabilities, the "usual" might actually be safer than "safer but different".
Since he seems mindful of his declining capabilities (he is restricting when and by how much he drives), i wouldn't worry too much, he would most likely stop driving on his own, just keep checking him.
> to claim that this is a meaningful restriction feels wrong on many levels.
You must be joking. Requiring me to have a technology in my car that can apply the brakes without any input from me and without warning? That is most definitely a meaningful restriction. You're forcing me to trust the government's estimate of the relative risk vs. benefit. To be blunt, the government sucks compared to me at estimating my relative risk and benefit. Plus, government institutions have repeatedly shown that they will lie to me about such things to serve political ends. So no, not buying it.
> I'm restricted such that I have to wear seatbelts. I can't buy a new car without rear view camera. I can't buy a new car without antilock brakes. Headlights. Brake lights. I could go for a long time on this list. Are you claiming none of those make driving safer?
The question is not whether these things make driving safer as compared to not having them.
The question is whether the government regulations we have, which include requiring these things but also include lots of other regulations that have far worse justifications (if they even have justifications at all and aren't just blatant attempts to extort the public, such as speed traps), are a better system, on net, than a system in which such things were a matter of voluntary choice, but we forced people to bear the consequences of their bad choices. Oh, you weren't wearing a seat belt and you got in an accident? Sorry, your insurance company will not cover the medical costs due to injuries you suffered that the seat belt would have prevented (or you have a much higher deductible in such a case). Oh, you hit something while backing up and you didn't put a rear view camera in your car? Same sort of answer from your insurance company. Oh, you hit something because you didn't have anti-lock brakes and so you couldn't steer to avoid the collision? Same answer.
In other words, if it is in fact true that these various technologies reduce the risk of accidents, then people who know they will have to take the consequences if they cause an accident will use these technologies to reduce that risk. But in our present regime, our society appears to be violently allergic to such things. The very existence of no fault auto insurance, for example, indicates that we don't like the idea of having to take the consequences of bad choices--but the result is that everyone pays more for auto insurance (IIRC, average rates in no fault states are about twice what they are in states that don't have no fault).
We would all be better off if we abandoned the idea that government regulation can magically protect people from bad consequences. We would also all be better off if we abandoned the idea that we can somehow magically have a government that only makes good regulations, and accepted the inconvenient truth that government regulations come as a package, which even if it contains some good regulations contains many more bad ones, and all things considered is a net cost to us all, not a net gain. We need to stop depending on the nanny state to protect us and be responsible adults who are capable of judging risks and acting accordingly.
> Consensus appears to be 300-400k people saved since 1975.
I don't see where that number is given in the article you referenced.
In any case, I didn't ask for a number, I asked for a percentage of people who didn't wear seat belts who suffered or caused harm because of it.
> I don't see where that number is given in the article you referenced.
It's not hard to find. https://www.bankrate.com/insurance/car/seat-belt-statistics/ is an example claiming 329,715.
> In any case, I didn't ask for a number, I asked for a percentage of people who didn't wear seat belts who suffered or caused harm because of it.
Yes, I did notice that, and the not so subtle sleight of hand involved in it.
Theoretically OP's concern is still possible with AEB but it would require non-trivial coordination between offensive drivers where each subsequent driver comes in at a lower speed than the previous, cuts OP's car off -- now slowed -- with the same distance as previous in order to force a standstill. It would be a neat stunt driving trick.
I find the lane keeping and departure warnings lessen the cognitive load on driving. I find I'm able to expand my focus when driving a bit more and improve my overall situational awareness. Overall, I'd say it's a great thing. That and 360 degree birds-eye view cameras when parking are safety features that all cars should have.
It's not FUD, they're not stating it as a fact, they're asking for an analysis of AEB in lots of different situations and not just overall.
That is how many FUD tactics are done. It is a rhetorical trick to try and maintain an exit...
> This conjures the fear that the agency in question is bad at this. Do you have evidence that they are?
Yes. This was my movivation for questioning the "data says" logic. The NHTSA has made decisions based on data that were poor decisions because they misunderstood the problem domain and became overly reliant on specific metrics. Here is a copy-paste of an earlier comment in this thread:
--
> the NHTSA is a pretty cautious organization, and they like numbers.
They may be cautious but they're not necessarily smart. Their vehicle safety ratings only measure the safety of occupants in that vehicle. Result? An arms race where cars keep getting bigger, heavier, and taller to do better in crash tests against bigger, heavier, and taller cars. These "safer" vehicles are now more deadly to everyone else, especially pedestrians and bicyclists.
McNamara fallacy in action.
I'm at against bigger cars as you can be. Literally biking to places when I can. I am not clear that this is the nhtsa's fault.
That said, I could cede this point if you can show evidence of it in these studies. The org is a bunch of people, so I grant they could make mistakes. Without specific concerns, this is the definition of fud. You aren't giving criticism of these studies and actions, but seeding doubt on their data.
> In a cost benefit, if the number drops 38%, you are going to have to work to show it isn't worth it. What would the number have to be?
Where did anyone say "isn't worth it"? The sentence I see is "The problem with numbers is that they don't tell the whole story."
Though in a more general sense, if someone does invent a system that has strong average safety gains, but makes the majority of people less safe... that's a hard sell.
Fair, if you are just pushing that we need to keep looking at the numbers and to be aware of some ways it can go bad, fine. I took the topic here to be that these studies were bad. In a known and showable way.
It isn't a binary choice, you can disable the function for situations it's not suited for. Nothing should be forcing us to enable harmful scenarios just because.
well, the law explicitly says mandatory. Defaults tend to stick.
Respectfully, you're taking an extremely naive stance to a topic where others are trying to bring nuance. I think if you listen carefully, you'd learn something that will help you not just on this topic but in your life.
Logical fallacies like the base rate fallacy are real, and they fool all of us frequently, myself included.
Bringing nuance is fine, but it is also an easy stalling tactic in progress. Clear improvements in safety are shown in the studies linked here. From an organization that has shown remarkable progress in safety over the years. And has been beset with such "nuance" since the beginning.
Literally the same arguments were and are made against seat belts. Often. And is why I am not too willing to give way here.
Now, if you are just saying, "I hope they keep their eyes on the data and are willing to keep improving," great. I agree. If you are trying to angle it that this is not a safety improvement, I will loudly disagree.
What if it being optional increases death statistics massively? Should we really be ideologically driven instead of relying on data and measurable effects?
What if free will leads to wars, crimes, deaths? Surely the world would be safer if we sacrificed our humanity for security!
I suspect the real issue is people using phones while driving, I know of at least one person who is dead from this. I rear ended a car when I was in my early 20s while on the phone, I se people driving trucks in narrow streets while on the phone in my area.
I don't like enforcing things like this as a band aid, it's fine to try improve a situation, but let's be honest, many people who are involved in these accidents are doing something with their phone.
Sure. What we know, though, is "stop doing that" doesn't work. That's why aviation, highway safety, healthcare, tech postmortems, etc. have discovered that "please ask the human to stop doing human things" isn't the approach to take. Instead, we try to engineer away the opportunities for human error to kill people. (Quite successfully, in a lot of cases; as an example, aviation safety in the developed world is overall an incredible success story for regulators.)
People will continue to use their phones in their cars. That's a fact we have to accept.
> the not so subtle sleight of hand
If you think that was sleight of hand, we obviously have very different attitudes about risk and regulation and we probably won't be able to have a productive discussion.
Yes, but sharing a characteristic with FUD is not enough to make something FUD. It's a realistic concern.
Someone else mentioned airbags. Airbags are great overall but increase danger for some people, and that's without obvious signs of anything being wrong.
With AEB we know it makes dangerous mistakes in significant amounts. So we very much should analyze different implementations under different kinds of weather, road, and driving style.
We don't know it makes dangerous mistakes. Randos on the internet assert that. The data shows fairly convincing evidence is reduces wrecks.
Is it a new behavior and jarring when engaged? Certainly. Should we keep watching the data for problems? Certainly. Should folks report all problems so that they are tracked? Certainly.
Do most people actually have trouble driving safely? Despite their insistent view of how great they are at driving? ...
I think it is probably overall a safety improvement, and I think many of the arguments about losing control of your car are a bit overblown. But, there's something to them, and it rubs me the wrong way to ignore and dismiss those you disagree with. It's too easy to characterize a position you don't agree with as one that is operating in bad faith or ignorance.
We can easily pass a law requiring AEB in new cars. Can you explain the mechanism that would allow legislating free will? Or were you just looking for any counter example, no matter if it makes the least bit sense?
You think statistics override basic principles of human dignity. I'm merely exposing your argument for the absurdity it is.
> What if it being optional increases death statistics massively?
What else would you like to make mandatory to satisfy your death reduction statistics, I wonder? Vaccines?
Let's just go all the way: you're talking about eliminating human agency, autonomy and free will. You want to replace it with decisions made by committee with optimized outcomes and make them mandatory with force of law.
> Should we really be ideologically driven instead of relying on data and measurable effects?
Yes. Do you not have principles, values you won't compromise on?
> That's a fact we have to accept.
You can go ahead and do that but I think people are justifiably pissed that we require invasive technology to coddle the idiots who can't be bothered.
I don't like the whole, "we'll protect you from yourself attitude".
Do I need to go find dashcams? We know it makes dangerous mistakes. Come on.
This is not one I will cede without evidence, so yeah. With the amusing extra burden that I have yet to know a safe driver that used a dash camera.
I have been in a car where the alert went off that it was engaging aeb. The driver was far too close to the car in front of us...
I'm definitely not ready to sacrifice my fellow humans on the altar of my own conviction. But hey, you do you!
Let's see, what can I find on youtube in a couple minutes of searching. In increasing order of wrongness:
https://www.youtube.com/watch?v=BLRRBHNjCn4 There's a car that it's reasonable to worry about, but it was clearly braking and almost stopped by the time AEB kicked in, and AEB kept going as that car went to zero.
https://www.youtube.com/watch?v=TO6tJnM_HSs Adaptive cruise was already in control and working fine.
https://www.youtube.com/watch?v=hCk6yrN453c Absolutely nothing in front.
And whatever happened here probably counts: https://theintercept.com/2023/01/10/tesla-crash-footage-auto...
511 Comments: