Hacker News Reader

I used to have a roommate (who has several accidents on his record) with the exact same model of Toyota Corolla that I had. We’d frequently go on trips together.

My driving experience: very rarely have a PCS audible warning when conditions change unexpectedly on the 10 (SoCal). Braking activated once or maybe twice in three years.

His experience: constant “false” alarms from the system, several instances of pedestrians and vehicles “coming out of nowhere” that passenger saw 10+ seconds away, several instances of slamming on the brakes, both machine and human-activated.

Some people just don’t treat driving as the life safety critical activity that it is. Most people.

I tend to agree, especially in the situation described in that comment. I have a 2020 Subaru Outback that I believe has a very similar braking system, and while I have also experienced a small handful of false positives (especially when moving very slowly in a tight parking spot), I think the system is pretty good, and I do feel good about its ability to help at least a little bit if I ever really need it.

I’m willing to ignore the false positives when moving very slowly in tight parking spaces. That’s startling and annoying, but doesn’t really bug me too much. I believe you can also turn that off independently of the forward-facing road-speed emergency braking feature (and you need to if you have a trailer or bike rack).

As for false positives while at road speed, I think the actual brakes have only kicked in once or twice, and only very briefly (certainly not enough to slow me down so much that it could cause an accident). It’s a bit more common to get a false positive on the first stage of alert, which in my car is an audible alert and some red LEDs on the primitive heads up display.

Anecdotally, I’ve found similar. Never had phantom breaking in my crosstrek. While it’s possible someone’s car could be miss calibrated, I’d guess pebkac is more likely and self reflection of driving habits is needed.

The only object warnings I’ve triggered are completely understandable and never had auto braking

A sharp turnning off-ramp to a rest stop on i5 with a sign in the middle that one drives directly at before turning has reliable triggered the object warning but never breaking. I suspect that if I went dangerously/stupidly fast in the turn it would brake.

The other other time I could reliably trigger it was pulling in faster than usual to park head on 1 ft in front of a fence, and that only ever did the warning not braking.

So far, and it’s anecdotal yes, I’ve never had it happen when actually on the road.

What I do find annoying is the cruise control starts to floor it when changing lanes if it sees the clear path down the lane line during the change.

My experience is largely the same as yours. I've had the alert (no braking) several times on the highway during sharp corners or when coming up on slower traffic (when I've been totally aware, but probably coming in hotter than I oughta be). I've also had it trigger twice when entering my apartment building's parking garage. It has a very steep uphill ramp and a couple of times I must have approached it fast enough that it thinks I'm running at a wall.

I also agree with the cruise control issue. The adaptive cruise control is very nice for highway traffic that's changing speeds a lot, except that I'll forget that it's set at 65-70 after cruising around 50 for a while, and then when it finds a clear lane it will try to accelerate very quickly. That's ultimately just a limitation in how far ahead the system looks to find cars.

That was just one example. I've clarified my original comment, previously I was using the term "car length" in a way that did not make sense to others. I should have said "seconds of follow distance"

All of this is infinitely nuanced anyway. There might not be a side street. It can happen on the highway too when cars switch lanes to exit. And it's not like it happens every day, it's fairly rare, I just have yet to actually _need_ the car to do it, by a fairly wide margin.

1-2 car lengths (not in literal units of 15-feet each, I mean one or two _seconds_ of follow distance minimum after accounting for speed/traffic/weather) going the speed limit with my foot off the gas and over the brake, paying attention ready to slow down fast in case of a contingency feels very safe to me. Again, pretty nuanced, this only lasts for a handful of seconds while the lead car completes their maneuver.

The salient point is that I already do what the machine tries to replicate, I don't need my maneuver clobbered by another system.

I’m not comfortable at all claiming that you’re a bad driver in general, like the GP proposes, and maybe I’m taking your unit of measurement too literally, but I don’t think 1-2 car lengths is a reasonable follow distance in almost any circumstances above 45mph, two lengths is kinda the minimum to not bother the person in front of you. Again, I’m making a very general statement, that might not apply to the situations you’re describing.

And I’m a young guy that is working on being a more conservative driver myself - so I’m chuckling to myself just at the idea of me taking this side of the discussion.

This is an interesting point. The Subaru's adaptive cruise control system treats a "car length" as a fairly large unit of measurement. One car length seems to be the literal length of the vehicle, plus some reasonable amount of buffer on both ends. Probably twice as large as the vehicle itself. And the faster you're going, the larger a single car length becomes. Infinite nuance :P

But in general, very good commentary on systems that provide useful feedback for the human operator versus ones that don't. ACC makes it clear what a car length should be, and if you're paying attention you can gauge how much more space it leaves when you're moving fast. So that's what I go by.

The Subaru uses two seconds per tick mark on the adaptive cruise display, so the distance changes as you go faster. At slow speeds, it's roughly exact car lengths, and at 65MPH it's considerably further.

Fwiw, it’s interesting to me that the unit of a “car length” would not refer to a literal car length (14-17 ft or so).

Have you ever looked up stopping distances?

https://www.automotive-fleet.com/driver-care/239402/driver-c...

I don’t know how accurate the above is, but it suggest that the stopping distance at 40mph is 5-9 car lengths, so I have to agree with the above poster that if you are routinely leaving one car length and your car system is freaking out, you are probably driving unsafely.

Keep in my that statistically this could work out for you fine for 20+ years before one day you get into accident, and presumably blaming the driver who suddenly stopped mid turn.

Check out watchpeopledie or idiotsincars for plenty of videos of what could go wrong. I recall one video of exactly this issue — an idiot SUV stopped mid turn in the middle of a lane, clearly a bad driver. And then another driver t-boned her because he didn’t have enough distance to stop, also cleary partially at fault IMO.

I know people with lifelong pain from minor auto accidents. Don’t have that be you, or cause that for someone else. What do you lose by keeping some more space between you and the next car?

One car length is way more than 15 feet to my mind, probably 40 feet or more depending on speed. See my other comments about how nuanced this entire dialog will inevitably be. Don't you think it's more reasonable to assume we have different definitions of a car length? Perhaps I've used the term in a grossly incorrect manner. "One car length" to me equals "one unit of minimum safe distance taking into account road conditions, traffic conditions, and speed"

Not just 15 feet, that'd be insane!

I think you are confused (either that or I am!). I’m pretty sure a car length is literally the length of a car, which for a Crosstrek is around 15 feet.

Given this confusion, it does seem like it explains the root of the disagreement here, but that said, I don’t think it’s unreasonable for people reading your comment to assume that when you said “car length” you meant the length of a car.

You could just admit that you used the wrong term? One car length = the length of a car.

Sure, I'm fine with admitting that. It was an inevitable misunderstanding anyway because cars are different lengths so "car length" is largely meaningless.

The Subaru's adaptive cruise control system treats car lengths as variable units based on speed. It's much more liberal than I made it sound in the original comment. It sounded reasonable when I wrote it, but infinite nuance strikes again I suppose.

Have an upvote for the adaptive cruise control comment. It's this way on the Tesla, too - follow distance is described as car lengths, but it's actually speed-parameterized follow distances in reality.

Personally, I find it somewhat confusing, but people should probably be more aware that it's a thing and not necessarily someone mixing up terminology.

> One car length is way more than 15 feet to my mind, probably 40 feet or more

Cars in your area are on average 40 feet long?

A car length is about the length of an average car.

Either way, with a 40 foot "car length", 1-2 car lengths is then ~80 feet. The above stopping distance chart puts a 40mph stopping distance with reaction time at almost 140 feet. <80 feet away is driving way too close at 40mph! Especially for a car you know is doing a maneuver which could result in them braking suddenly such as turning into another field of traffic.

With the amount of semi-trucks in my area, actually yes, average length of vehicles you encounter on the road is probably close to that.

I just used the term "car length" with entirely too much room for interpretation and am now getting torn apart for it, my mistake. I swear I'm not tailgating people LOL

> I swear I'm not tailgating people LOL

If you're going 40mph and approaching nearly stopped cars at <80 feet, you're absolutely tailgating people and creating extremely risky situations.

Don't get me wrong, I've had my emergency braking trigger for similar circumstances of someone turning into a parking lot off a feeder of a highway. But I don't blame my equipment, I acknowledge I'm not a perfect driver. The car knew it could have been quite a challenge to stop in time and so it took action. The better action would have been for me to be going slower and approach that situation with less speed, not to just blame the car for being cautious and forcing realistic safety margins.

Nobody's approaching nearly stopped cars, I think you have misunderstood some other facet of my post. I sometimes approach _decelerating_ vehicles, _while I am also decelerating_

> Nobody's approaching nearly stopped cars

Yes they are, further up this comment chain is:

> for example, if they're turning and will be gone in a second anyway

That car turning is a nearly stopped car, doing a maneuver which could result in them stopping suddenly and unexpectedly. Going 40mph less than 80 feet away from that car turning is absolutely an extremely risky manuver. You don't know they're going to be gone in a second, you're assuming it, and if your assumption is wrong its a collision because you cannot possibly stop in time. Emergency braking triggering in this scenario is 100% justified.

You can turn the wheel of a vehicle without being stopped. People change lanes at speed on the highway every day. Please try to assume good faith. "Turn" does not always imply a 90 degree turn with a full stop prior. I am not approaching cars at speed while they are stopping to make a 90 degree turn...

Changing lanes is called changing lanes, not turning. They're inherently different maneuvers.

Either way, if you are rapidly closing the distance to a car which is changing lanes, you don't know for certain they won't cancel the maneuver and go back to being in the lane. You're still assuming that, and assuming you'll be able to slow or stop fast enough to handle it.

If your AEB is constantly getting "false" alarms, maybe it is not the AEB that needs adjustment. Maybe driving in a fashion where the car's AEB doesn't constantly give "false" alarms would be a lot safer.

It is by no means a constant occurrence. I've edited my original comment to disclaim this as you seem unable to assume good faith about any of this. It has happened only twice, both times with a fairly wide margin where human intervention could have occurred should it have been needed (it wasn't).

Changing lanes is a type of turn because you _turn_ the wheel to change lanes. You also might turn 45 degrees in a roundabout, or 90 degrees at a stoplight. They are all turns.

I'd agree that it's an assumption. Though your assumption that the distance is closed _rapidly_ is also just that - an assumption, and one made in bad faith despite my efforts to correct it.

The mere act of being in a car on the road involves dozens of assumptions made every single minute about the traffic conditions. Very little is known for certain, ALL maneuvers are to some extent based on assumptions.

This conversation is no longer productive, thanks for your participation anyway though.

> One car length is way more than 15 feet to my mind, probably 40 feet or more depending on speed.

A car length is, literally, the length of a car. It does not vary with speed.

Exception: https://en.wikipedia.org/wiki/Length_contraction (coming soon in a software update to a Tesla near you)

Sure, but saying this is meaningless when we're actually talking about follow distance. Cars are different lengths.

To me, a car length should simply mean the minimum safe _length_ at which to follow the car in front of you.

This would lose all communicative benefits. Teaching a young inexperienced driver “just always keep one car length” where one = safe, is meaningless.

I don’t know you, but I believe a majority of people underestimate safe stopping distance. That’s why the webpage I posted, which is one of many similar ones, exist. That’s also in part why there are so many accidents.

People are also bad at estimating feet, so telling someone keep 80 feet is also not helpful.

But people are pretty good at estimating how many cars can fit between them and the next car.

So telling someone, hey, you are tailgating that guy and only one car could fit in front of you, but you need 10 car lengths to be safe, has a lot more communicative power.

> Teaching a young inexperienced driver “just always keep one car length” where one = safe, is meaningless.

Not if taught correctly, which I think it's important to note that we can't really do via textual communication.

In the driver's seat, during a driver's education course, you might learn that a car length is the length of an average car, but you also learn how to count the seconds of stopping distance you have in front of you. In addition, you learn that weather, traffic, and other factors can affect stopping power/distance, the point being that the whole system is incredibly dynamic and variable.

One can't reduce a dynamic traffic situation down to "X car lengths = safe." Similarly, saying "1 car length = a safe distance" doesn't really help either.

The idea is to teach people what "safety" actually looks like. How it's referred to is secondary. The proof is in how badly I was misunderstood when I said 1-2 car lengths. I _know_ I'm keeping enough distance, because I know how to judge a safe traffic situation.

I don't really care what a car length is or isn't, it's fairly irrelevant because safety is a much bigger picture than units of measurement.

Exactly; car lengths vary by model, not by velocity.

I get it. You'd prefer to use the phrase "car length" to not literally mean "length of a car" in this context. Apologies if I seemed pedantic but, as evidenced by this thread, your use of car length is going against the grain.

Safe following distance is often expressed in multiples of car length. What car? Sure, good point. Pick one; it's close enough! (If you're actually thinking about following distance in any terms at all, even roughly estimated, you're ahead of the safety curve and you clearly are!)

Ideally, following distance could be accurately estimated in metres and adjusted continuously by velocity and driving conditions. AI can do that. People just kind of eye it and say, "it was raining and I followed about four car lengths behind him" or "we were inching along and I was half a car length behind him". Good enough for meat brains!

Safe driving!

> And then another driver t-boned her because he didn’t have enough distance to stop, also cleary partially at fault IMO.

Not partially at fault; completely and totally at fault (except if we're talking about someone deliberately trying to cause a crash in which case I'll settle for partially). The reality is there could be a hundred reasons the car stopped, the driver being an idiot is only one of them. Mechanical failure, something unseen from the following car, a door flying open, someone stepping out in front of them, a queue in front of them which is not obvious from behind etc.

If you are not leaving enough room to come to a stop when the car in front does then you are riding your luck, there are no two ways about it.

It sounds like the example was not of a car they were following, but one that pulled out infront of them and stopped in their lane. Just as you are referencing possible excuses for that driver, we could come up with possible excuses for the driver that pulled out (like pulling out suddenly without enough distance to react even if the other car was going under the speed limit).

We also need to understand that being at fault isn't 1-to-1 with who was being safe. Typically, if you pull out and get t-boned by someone who has the right of way and is generally following the law, the person being t-boned is at fault for failing to yield. Where a vehicle is impacted plays a huge role in investigators determining who is at fault. Now if you run into the back of the person in front of you, it's almost always your fault, unless the person in front was driving carelessly or recklessly (brake check, for example) but would still involve partial fault in many cases.

So really, we can't say either way without knowing a lot more.

It's effectively impossible to maintain safe stopping distances in many places. Speaking from experience in Chicago, it's just a concept that doesn't exist for many people. And if you are driving slower in order to maintain a safe stopping distance, you're creating an extra hazard on the road by maintaining a larger speed differential from the rest of traffic. Not to mention, you increase your own risk of accidents because of all the people that will aggressively cut you off.

I personally always try to maintain a safe-ish stopping distance, but I'll also ensure that I'm never in a position where I'm missing a swerve ready escape route.

Little known fact: You can travel the exact speed of the car in front of you while also being any distance N behind them. If I'm traveling greater than 50MPH, I leave at least 3 car lengths between me and them, this is to allow for reaction time to stop at the same rate they are stopping.

In California that just means that in the span of 15 seconds 3 other cars are going to move in front of you and the car that was ahead of you.

Why does that matter at all? Just slow down to get more space whenever that happens.

In my experience they will aggressively cut you off anyway, so you might as well give yourself some breathing room.

I don't understand this - are you implying that the only way to safely drive on highways is to always maintain a follow distance equal or greater to your stopping distance?

I don't think it's reasonable to prepare for the car in front of you to stop in place as though it had hit a brick wall. Instead, the safe follow distance is derived from the current speed and breaking power of the vehicle I'm operating and the vehichle that's behind me, and the distance to the vehichle behind me.

2 car lengths? So like, 250 milliseconds at 40 miles per hour?

100%. We drive Subarus, highway and city. Have NEVER had an emergency brake triggered. OP is unsafe.

"And I'm 100% in favor of systems that force other drivers on the road to drive more safely as well."

Ug. I'd rather have testing be changed so that safety is increased in all situations by having a competent driver. Most of the situations being described here are simply poor choices on the part of one or more drivers.

For real, who the hell is cutting through lanes of traffic and slamming on brakes because they will miss a turn? Best to remove these people than have them rely on a price of tech to save them in 1 situation but let them continue to be dangerous in another 9 areas.

>Ug. I'd rather have testing be changed so that safety is increased in all situations by having a competent driver.

This follows to taking driving license from them.

You'd have to change USA more than Lincoln and Roosevelt together to remove critical dependency on cars there.

> You don't know if the car in front of you making that turn is going to brake suddenly

More drivers need to be aware of this one. There's an intersection near me where drivers always tailgate and rush others through the (slightly greater than 90-degree) turn onto a side street, no matter how well the turn has been signaled already. There have been multiple minor accidents, but fortunately no pedestrians have been hit that I know of. When someone's turning, drivers behind need to give them more space not less.

Maybe? Few would openly admit to be a bad driver to begin with so it is hard to even get anecdata on the subject. In other words, I am not sure you can determine an answer either.

What I can say is what bothers me about it. Can I turn it off? Can toggle on/off switch be disabled by an update? Can I trust that it the off button will do what it is supposed to do? If I can do those things, can I trust this feature to work as advertised? I used to not ask questions like that with cars as there was certain level of.. expectation that this machine has to work in sometimes drastically radical conditions ( especially in some parts of US ).

Cars are software now. But that is not my main beef with it. My beef is that it is built to the detriment of the end user and to strip the user of all control.

> You really shouldn't be driving like that. You don't know if the car in front of you making that turn is going to brake suddenly, like because, I dunno, they suddenly see a pedestrian crossing on the side street.

I think that's the problem with that attitude and why I don't like my car doing excessive braking. Unlike the distance lidar, I can actually see there's nothing obstructing the turning driver, and I can see their speed of leaving the lane is simply too high for them to stop before leaving the lane even if they slam the brakes. And even if they did somehow manage to break where they are already I'd only need a little swerve to the left to drive around with no traffic there etc