The Insurance Institute for Highway Safety just released the results of two month-long studies they undertook to determine how drivers interact with semi-automated driving systems, specifically Volvo’s Pilot Assist and Tesla’s Autopilot. Both are Level 2 semi-automated driver assist systems, which means they require constant human driver attention even as the automated system handles a significant portion of the driving task. Unsurprisingly for anyone who has ever met a human being, the studies found that drivers that use these systems are far more likely to focus on things other than driving and are more likely to be distracted. Because no shit. Of course they are.
For each study, a group of drivers (14 for the Tesla study, 29 for the Volvo one) was given access to cars with semi-automated driving systems; these drivers did not have previous experience with such systems and drove these cars using the semi-automated assist systems whenever possible for a period of a month. The big takeaway was that when using these systems, both of which have a number of safeguards in place to prevent driver inattention from the road, what people seemed to be especially good at was learning how the driver attention systems nagged and tried to get attention, and how to get around those safeguards so they could focus on non-driving tasks anyway.
You know, like people do.
Here’s what the results of the Volvo study had to say:
Results: Participants exhibited higher odds of visual-manual distraction or driving hands-free in period 2 when using Pilot Assist relative to manual driving, but patterns differed noticeably across the groups. Pilot Assist use among groups A and B was associated with higher odds in the second period relative to the first, whereas group C exhibited a high level of visual-manual distraction and hands-free driving when using Pilot Assist throughout the 4 weeks of data collection.
Discussion: Our findings suggest that drivers show reliably less attention to the road with the versions of partial automation we tested compared with manual driving that develops with time or is evident relatively early during drivers’ initial trips with the automation. The results support arguments for driver-monitoring solutions that ensure adequate attention to the road. Differences among the samples in patterns of behavior change highlight the need to study factors that may further modify how drivers adapt their behavior when using partial automation such as driving exposure, willingness to use automation, and iterations of partial automation that differ in functional performance.
… and here’s the Tesla results:
Results: We found that drivers learn to internalize safeguard sequences and discover windows of opportunity to do non-driving-related activities. People learned to respond quicker to alerts, leading to fewer escalated sequences in the latter half of the study. However, drivers also spent more time engaging in non-driving-related activities and glancing off-road, which corresponded with more initial alerts of the attention reminder sequence as time went on. Prolonged disengagement culminated in 16 lockouts across the sample, although in general, drivers responded faster and had fewer lockouts over time.
Conclusion: Our findings demonstrate the human ability to learn system constraints and thus illustrate that it is possible to shape safer driving behavior with robust safeguards. User-centric design considerations for driver support strategies are presented in this paper.
In some ways, the conclusions here make me perversely proud to be a human being. You give us rules and constraints and tell us there’s something we can’t do, and you can bet your ass we’ll figure out some way around it. Of course, we often do this even when it directly impairs or endangers our own well-being, as it definitely does here, in the case of paying attention while a semi-automated driving system is in control of your car speeding along at a mile-a-minute.
The humans behind the wheels of both cars found ways to engage in distracted, non-driving behaviors. The Volvo study found that drivers were distracted 30% of the time, which is alarmingly high, and in the Tesla study, over the course of 12,000 total miles, almost 4,000 driver-attention warnings were given, which is equally alarming.
All of this just reinforces a belief I’ve held for a long time: the issues around Level 2 semi-automated driving systems – pretty much the only type of systems available on the market today (unless you count the few Level 3 systems, which are even worse) – are inherently flawed not because of technology reasons, but because of how humans work.
If a human is in a system where some automated machinery is doing most of the work and their job is to monitor things, they will get distracted. This has been well-studied and understood since Mackworth’s famous 1948 study “The Breakdown of Vigilance during Prolonged Visual Search” which demonstrated that intense and prolonged vigilance to a mostly independently-operating system is simply not something people are good at. Monitoring a car that’s mostly driving itself is a task like this, and the danger here is that semi-automated driving systems are by no means perfect. Even though they’re very impressive and getting better and better, they still make mistakes, which is why we require a human to pay attention in the first place.
Sometimes it’s not entirely the fault of the distracted drivers, though; a shocking number of users of automated driving systems think their cars are far more capable of driving on their own than they really are.
Both studies found that drivers will learn to do whatever a semi-automated driving system requires them to do to “prove” they’re paying attention, whether it’s nudging the steering wheel or keeping eyes pointed in a certain direction, but they just learn what the system wants to let them continue to engage in non-driving, distracted activities, like ones noted in the study which include grooming, eating, and, of course, looking at phones.
This could mean that more robust driver-monitoring systems are needed, though my money is always going to be on humans figuring out ways around these more robust systems. Fundamentally, I think Level 2 systems are just incompatible with how human beings actually, realistically operate, and efforts would be better spent developing robust Level 4 geofenced automated driving areas where they’re needed most.
We’ve basically heard all of this before, and still nothing has really been done. I wonder how many more similar IIHS studies we’ll see with essentially the same conclusions before anyone decides any of this matters.
It’s not you, tech. It’s us. We’re the problem.
The Feds Finally Crack Down On Tesla’s ‘Full Self-Driving’ Because It Breaks Traffic Laws Just Like You Do
These Tesla FSD Hands-Free Mod Devices Are Such A Bad Idea And Promoting Them Is Dumb
It Turns Out That Some Advanced Driver Assistance Systems Might Not Actually Reduce Crashes
New IIHS Study Confirms What We Suspected About Tesla’s Autopilot And Other Level 2 Driver Assist Systems: People Are Dangerously Confused
I dunno I tried the blue cruise on my Mach E for the first time on a long trip this weekend and I liked it a lot. I mostly kept one have in the wheel because I was a little unfamiliar and paranoid, but boy did it help a lot on a long, tired ride home on the interstate?
The thing would bug me pretty often if I looked at the scenery too much.
Interesting that this study did not include Ford models. The Bluecruise on our Mach E is impossible to fool in my experience. I looked out the driver door window for about 5 seconds and the car got mad at me for not looking out the windshield and threatened to turn the system off.
If we’re going to have one charging standard, perhaps there should also be one self-driving standard as well.
No surprise by the number of two hand texting Tesla drivers I’ve seen.
My ioniq 5 has Highway Drive Assist. It drives like it’s put back 2 or 3 beers. I only use it on late night empty freeway or stop and go traffic. Absolute sanity saver in stop and go, and does the task amazingly well.
I see people looking down at their phones in pretty much all cars on my way to work. At least if they have autopilot or something like it the car will brake and maybe hold lane position.
I don’t think people pay attention when driving a damn automatic let alone autopilot bs.
This isn’t good when Teslas are programmed to break laws…
That’s FSD supervised and not autopilot.
I think the point still stands – driver assist software that is designed to help a driver break the rules of the road is not helping the situation at all.
I know I am a less attentive driver when I drive a car with an automatic gearbox vs. when driving a manual. The more half-assed automation added to cars, the more half-assed the driving. Level 5 or die, literally.
I agree, if we’re going to automate driving then it has to be full on level 5. I want that shit tested and proven before its allowed on public roads. None of this level 1-4 bs.
So if you ask these companies what the advantage of level-2 semi-automatic driving is, what is their actual answer? What possible benefit could automated driving have if not “you can do other things besides drive”? So if you can’t say that about your automated system, it’s just pointless. I suppose the idea is that it’s a necessary stepping stone to full self-driving, but then they’re just having their customers fund their R&D… which is good for the ol’ bottom line, I guess.
Anyway, if you’re sitting there with your hands on the wheel and looking at the road, is it really a huge benefit over just, you know, driving? If you’re too weak to turn a powered steering wheel with your own power, maybe it’s time to just give up driving (and/or do some exercises).
I drive with one and it reduces fatigue on long road trips. I don’t feel as tired when I get to my destination which is probably better across the drive.
Find me one situation where a human actually delivers 100% interaction? Not including sex or masturbation.
Wait? You are supposed to focus during sex? Damn. I’ve been doing it wrong.
Yes. Focus on baseball
“….and we do it doggy style so we can both watch X-Files….”
-The Bloodhound Gang
“Beige.”
“What?”
“Beige – I think we should paint the ceiling beige.”
I know that is easy to be dismissive about this study. You can easily fall into the “no shit sherlock” kind of reaction. I think these knee-jerk reactions have become all the more common in the internet age while actually ignoring the significance of this report.
Studies like this are extremely important for actuaries to build insurance models. Insurance companies on an annual basis (typically), have to request rate increases to various provincial and state insurance boards. The insurance boards have the final say on whether to approve these increases. Boards require hard data like this report when assessing whether the requested increases are justified and get approved.
I have a feeling cars with these level 2/3 autonomous systems are going to see large insurance increases given the now scientifically documented danger of level 2/3 systems.
So I for one am pleased to see this report was done and showing the dangerous behaviour that occurs when using such “autonomous” systems. If people want to buy and use these systems, they deserve to pay more for the risks that are created.
The problem is many people will be punished with higher rates for a system they may not use/don’t actually want just because of how car buying is today. Do you want a sunroof and cooled seats? Well you gotta step up to the Luxury trim bundled with the 7,000 watt sound system, Safety Net Pro+ and all weather floor mats. I turned it all off in my cars because it causes more annoyance and distraction than driving when I’m in traffic. I’m constantly wiggling the wheel because it bongs at me every 5 seconds for not moving the wheel even though I’m on a dead straight section of highway doing 3mph. The point of the system is “traffic jam assist” yet it gets mad when you don’t move the wheel because you’re in a traffic jam.
That is definitely an issue. But so far (in Canada at least) the advanced driver assist is a separate very expensive charge item for most EVs. I’m not sure about gas fueled car since I’m not interested. (12 years this December driving electric will do that to you.)
I suspect that the ADAS systems are an option so the manufacturers can get the MSRP below a level so they qualify for a $4500 government rebate. For example, the Ford Lighting has the high capacity battery as an option on all trims so they can keep the MSRP low enough to qualify for the rebate.
I’m just here for the Del Taco taco.
Disappointed that I missed that article.
2nd level systems? Talk to me again when they complete graduate school.
I heard a story about this study on the radio in the car this morning and said out loud to no one, “they needed a study for that!?”
Perhaps the automakers need to up the ante with the warnings attached to 48V electrodes in the driver’s seat? They could move beyond the realm of annoying and redo the study.
Sort of related, back in college I volunteered for a driving simulator study that entailed trying to use a calculator while driving, with a survey after of what types of warnings were noticed. The audible warnings and such I noticed, but I didn’t notice the seat/steering wheel vibration warnings at all. I told them the vibration warnings weren’t a good idea for anyone used to driving an old jeep.
♫ BURning DOWN the car!!!!♪
David Tracy does not get this reference.
You may find yourself behind the wheel of a large automobile. And you may ask yourself; how do I work this?
Where does that highway go to?
Am I right, am I wrong?
My God, what have I done?
I will not even Once in a Lifetime buy into this crap.
Nice!
I’m maybe the other way around. I see instances of the tech being wholly unable to safely know what’s going on around me to trust it. I don’t use autopilot. Just the adaptive cruise, and even that I use less and less because autopilot will step in and do something when it has no business doing something. In the last year, my Model3 has:
When I read stories of people doing other things while driving with their autonomous system on it terrifies me. They might as well be driving with a brick on the pedal and bungee cord holding the wheel. It’s simply not ready, and the degree to which it’s not ready is stark. Maybe highways elsewhere are more predictable? Maybe other people are ridiculously lucky? Next car, I just want plain old cruise control.
Wait? A tesla driver that is actually honest about your cars shortcomings? You are one in a million. Please hang around here forever
Maybe he bought it used and the depreciation had already hit and he wasn’t giving money to Musk.
I used to work in a factory full of tiny semi-automatic welders that were loaded with parts several times a minute by operators.
While I was there they introduced mandatory safety glasses for everyone. Several of the operators refused to wear safety glasses, because people hate being told to do things.
Things escalated until they were told if they didn’t wear them they’d get fired.
This ring leader for the resistance was suddenly quiet, head down, wearing her glasses.
The threat worked right?
Nope. She’d taken the lenses out of the frames and didn’t want to get caught. All of the grief of compliance but none of the safety.
People will not do what they are supposed to, regardless of how stupid it is to not do it.
“Nope. She’d taken the lenses out of the frames and didn’t want to get caught. All of the grief of compliance but none of the safety.
People will not do what they are supposed to, regardless of how stupid it is to not do it.”
As if Covid wasn’t all the proof anyone would have needed.
During an internship in the Ford Dagenham body plant back in the Cortina era, when spot-welding was done by hand, many of the spot-welders worked without goggles. Instead, they would turn their heads away to avoid the sparks for each weld, a few times a minute, all day long. Great for quality welds, and tremendous neck-ache?
It seems obvious that Level 0,1,2 systems should be banned since they are incompatible with humans. Or at the very least, ban auto-steering. And FFS, anyone can change lanes, the car doesn’t need to do it for you.
This is sort of tangential to this discussion, but I had a more recent car that had auto-emergency braking. And then I sold that car and bought an old ass car that didn’t have that. I won’t say I ever relied on the AEB system in my newer car, but it was one of those things in the back of your mind like “I have a bit of a safety net” in case something happens that I’m not prepared for. Now that I am driving my old car that has none of these systems, I find myself much more vigilant because I know my car will not save my ass. And it weighs a lot more than the old car too, so braking distances have to be adjusted mentally.
There’s nothing quite as unsettling as a malfunctioning ABS system. GMT400’s are notorious for it, usually wheel bearings eating the wheel speed sensors and triggering ABS at sub 4mph. Feels like trying to stop on ice in the middle of summer.
Pretty sure this was originally published in “The Journal of Advanced No Duh – v23 Iss.3”
I see dead people.
Whenever I get a rental with adaptive cruise control or lane keep assist that I can’t turn off, I am always paying attention while driving because I am constantly fighting the stupid systems. I’m sure there are systems out there that don’t ping-pong from line-to-line or lock up the brakes whenever it sees a road sign on a gentle bend, but I haven’t experienced those good systems – just the ones that do the dumb things I just mentioned.
The wife’s car has both of those and I’ll use them occasionally but I swear I pay more attention to the road when they’re on than when they’re off specifically because I don’t trust them and know it’s still me who will be skinned alive if I let one of those systems wreck her car.
I had a Dodge Hornet as a dealer loaner that had some of these features, and for the life of me, I could not figure out how to turn off adaptive cruise. It was infuriating to get slowed way down by the person in front of me who was slowing down on the off-ramp while I was continuing on the expressway.
As a tangentially related note, I’ve started noticing road signs that would confuse a vehicle’s camera system. An example would be a speed limit sign on a free-way siding that is not relevant for free-way uses. Another example would be all the times Google Maps declares that the next intersection has a stop sign, even though it doesn’t. Now that California is mandating a speed warning, all the times that a vehicle gets a road sign wrong will be yet another infuriating aspect of driving a modern car.
I don’t mean to sound like an old man yelling at the clouds, but it’s disappointing to see features that the automotive industry collectively had worked out regressing back to a worse state. (I.e., horrible cruise control, which, in fact, does not control your speed, or missing oil dip-sticks, etc…)
They popped like grapes?
Too bad I can’t read it, that article must be only be for taut sheepskin members.
These issues fall under the heading of PICNIC: Problem In Chair, Not In Computer.
There is another category called PEBKAC – Problem Exists Between Keyboard And Chair – but there’s no keyboard as such so maybe we can use Killing-machine instead. Anyone who unhappy with the use of the word Chair in this context may replace it with Cockpit.
The more cynical among us may assume these are simply ID-ten-T errors, which could very well be correct.
Time to create a new category. PEBSWAC. Problem exists between steering wheel and chair.
Solution: Torque Wrench (to fix the loose nut behind the wheel)
Not surprising as we also can’t be bothered to reach down to put on our own shoes!
But we’ll have MINUTES of free time at the end of our lives wearing those things!