Catch up on stories from the past week (and beyond) at the Slashdot story archive

 



Forgot your password?
typodupeerror
×
Transportation

Drivers Are 24% More Likely To Speed When Using Adaptive Cruise Control (arstechnica.com) 139

An anonymous reader quotes a report from Ars Technica: Back in the mid-'90s, the National Highway Traffic Safety Administration evaluated [adaptive cruise control (ACC)], logging 35,000 test miles (56,000 km). [ACC uses forward-looking radar to control the distance to a vehicle in front of it.] The NHTSA found that "ACC is remarkably attractive to most drivers. The research indicates that, because ACC is so pleasing, people tend to utilize it over a broad range of conditions and to adopt tactics that prolong the time span of each continuous engagement." However, the agency's results also suggested "that ACC usage has induced some elevation in the speeds that would otherwise prevail in conventional (i.e., manual and [conventional cruise control]) driving."

To test how ACC affected speeding, the IIHS recruited 40 drivers in the Boston area, then gave half of them a 2017 Volvo S90 to use for about four weeks and the other half a 2016 Range Rover Evoque, both of which were equipped with ACC. The vehicles were equipped with a monitoring system that included a video camera facing the main instrument display (to determine when ACC was operating), as well as vehicle speed and location, and the speed limits. IIHS also only recorded data on controlled-access highways. Analyzing the data showed that drivers in both the Volvo and Range Rover were significantly more likely (95 percent) to exceed the posted speed limit when using ACC than not (77 percent), although there were no significant differences between the two groups.

However, the absolute differences in speed were not that great. When driving manually, drivers averaged 6.1 mph (9.8 km/h) over the speed limit. When using ACC, this increased to 7 mph (11.2 km/h), or 7.1 mph (11.4 km/h) when using Pilot Assist in the Volvo. Interestingly, drivers sped more on highways with 55 mph and 60 mph speed limits than on 65 mph-limited roads. The IIHS estimates that "[c]ompared with manual driving, the increase in speed associated with ACC/Pilot Assist use was estimated to increase crash risk by 10 percent for fatal crashes, by 4 percent for injury crashes, and by 3 percent for property-damage-only crashes." The IIHS does note that it did not take into account the following distances while using ACC, which drivers can control (in increments of 1 and 5 mph). It also noted that drivers generally drive 5-10 mph (8-16 km/h) faster than the posted limit, and that's easier to accomplish when using ACC than when driving manually.

This discussion has been archived. No new comments can be posted.

Drivers Are 24% More Likely To Speed When Using Adaptive Cruise Control

Comments Filter:
  • by Penguinoflight ( 517245 ) on Friday March 12, 2021 @11:59PM (#61153096) Journal

    are the problem. Too much differential and people can't maneuver out of a bad situation, too little and there's inadequate flow within traffic. Adaptive cruise reduces the speed differential, and optimal increases the approach time of catching a slower car.

    I'd be more worried about people beginning to trust driving aids like this at the expense of attention, rather than criticizing them for imaginary problems like slightly elevated speed.

    • I’m guilty of using my AutoPilot to significantly reduce the mental energy required to drive. Sensor limitations are painflully obvious to me for the system, along with lack of a useful physics engine. As much as I hate driving, I am a better driver than my Autopilot.

      That said, driving takes so much of my mental energy that losing focus poses a significant hazard. Autopilot might only be useful real-time (rather than in anticipation of situations), but it doesn’t get sleepy or bored doing its

      • Dunno how AutoPilot works, but I find adaptave cruise control to be more taxing than regular. The last car I used it in had 3 distance settings; too close, way too close and OMGWTFWHYARE YOU SO CLOSE. Naturally the last one was the default setting.

        On an open road, by the time i'd be close enough for the adaptive adjustment would kick in I've already entirely disengaged cruise control to maintain distance. In stop-and-go traffic it might be nice, but how many models have adaptave cruise control that goes all

        • by dcw3 ( 649211 )

          Dunno how AutoPilot works, but I find adaptave cruise control to be more taxing than regular. The last car I used it in had 3 distance settings; too close, way too close and OMGWTFWHYARE YOU SO CLOSE. Naturally the last one was the default setting

          I've had an Infinity FX45 and a Dodge Charger, both with ACC, and no such problem. In fact, I usually use the shortest distance setting, but I'm also actively paying attention in case I need to intervene. I'm curious what you were driving?

          • I have personally seen similarly good results. Infiniti, Dodge, BMW, Audi, Kia, and Tesla all have very good ACC. Willing to bet all the Japanese companies are doing it right now. Once you understand how it works in your car, it can be great in the right situations. Tesla’s is the best, especially when not giving phantom braking in those “special” problem spots.

            • by Shaitan ( 22585 )

              I can add Honda to the list. I've used it driving across the country a few times.

            • by dcw3 ( 649211 )

              Yes, and mine were much older models...2003 Infiniti and 2012 Dodge...same with my wife's 2012 Jeep Grand Cherokee.

            • Yeah, the key is “once you understand how it works.” Before I bought a Tesla, I only ever rented cars, and I had trouble with adaptive cruise control. I think Tesla manages to do a little better/go a little further than most other systems, but maybe not quite as far as they make it seem.

        • The last car I used it in had 3 distance settings; too close, way too close and OMGWTFWHYARE YOU SO CLOSE. Naturally the last one was the default setting.

          On an open road, by the time i'd be close enough for the adaptive adjustment would kick in I've already entirely disengaged cruise control to maintain distance. In stop-and-go traffic it might be nice, but how many models have adaptave cruise control that goes all the way down to zero and stays engaged?

          Could you be more specific what kind of car gave you this experience? My Subaru's ACC default follow along is far enough behind that someone is frequently cutting in front of me due to the gap. Also, it goes all the way down to zero though a mild tap on the gas is needed to let it know you're ready for it to resume if it stops more than a couple of seconds.

          • I agree with this ACC assessment. I keep mine on the 2-of-4 setting for following distance. And it's just far enough that some aggressive drivers think it's a merging space, and I think the tolerance is enough for those same drivers to want to pass, thinking I'm "too slow" when really they're just over-reacting to the car's normal speed up and distance tolerance.

        • >Dunno how AutoPilot works, but I find adaptave cruise control to be more taxing than regular. The last car I used it in had 3 distance settings; too close, way too close and OMGWTFWHYARE YOU SO CLOSE. Naturally the last one was the default setting.

          The following distance option in the Tesla went away. So now it's following distance it too close and I invariably take over when it starts following.

        • Dunno how AutoPilot works, but I find adaptave cruise control to be more taxing than regular. The last car I used it in had 3 distance settings; too close, way too close and OMGWTFWHYARE YOU SO CLOSE. Naturally the last one was the default setting.

          This is probably because the system is simply not capable of maintaining a safe following distance... it doesn't work until you get too close. That's an example of a system which is unsafe by design.

          I don't know if all systems work that way, though, nor do I have any basis for comparison since you didn't say what kind of system you were using.

      • driving takes so much of my mental energy

        Username checks out.

    • I'd be more worried about people beginning to trust driving aids like this at the expense of attention

      If accident rates go down, why does it matter if people pay attention?

      Isn't the point of technology to allow us to focus our attention on other things?

      • Because then people will forget about the basic skills! Just like those pernicious books, downplaying the basic skills of memorization:

        If men learn this, it will implant forgetfulness in their souls; they will cease to exercise memory because they rely on that which is written, calling things to remembrance no longer from within themselves, but by means of external marks. What you have discovered is a recipe not for memory, but for reminder. And it is no true wisdom that you offer your disciples, but only its semblance, for by telling them of many things without teaching them you will make them seem to know much, while for the most part they know nothing, and as men filled, not with wisdom, but with the conceit of wisdom, they will be a burden to their fellows.

  • by EmoryM ( 2726097 ) on Saturday March 13, 2021 @12:00AM (#61153098)
    If I obeyed the posted speed limit I'd frequently be driving 5-10 miles per hour slower than the vehicles around me - I don't consider that safer than speeding.
    • by AmiMoJo ( 196126 )

      Part of the problem is that speedometers are not all that accurate. In fact some manufacturers deliberately make them over-read by quite big margins. For example the Nissan Leaf reads about 10% higher, so if you want to cruse at 70 MPH it needs to read 77 on the dash. If you set cruse control to 70 you are probably doing about 63 in reality.

      That means that even if you have a bunch of drivers all driving at the speed limit, because their speedos all have different readings some will be going several MPH fast

    • by hey! ( 33014 )

      I agree, although it does pose a dilemma in bad driving conditions. People don't adapt their speed and braking distance to things like ice and snow sufficiently, which is why there are more pile-ups in these conditions, which should not *necessarily* be the case. In good conditions I'll go with the flow even when that's 20 mph above the posted speed limit, but in bad conditions I'll be to the right going slower than many people, even though that's not ideal.

      • In good conditions I'll go with the flow even when that's 20 mph above the posted speed limit, but in bad conditions I'll be to the right going slower than many people, even though that's not ideal.

        The problem really isn't the speed differential. The problem is that people are dickfaces.

        Yesterday I made a six hour round trip drive to get stuff out of a storage unit. I went 70 down and 65 back, which means I had people passing me both ways. And motherfuckers are downright aggressive when you're not going as fast as they want. They wait until they're a half-second behind you to get into the passing lane, and then merge back in front of you only a half-second ahead of you, even when there's nobody trying

    • If I obeyed the posted speed limit I'd frequently be driving 5-10 miles per hour slower than the vehicles around me - I don't consider that safer than speeding.

      I noticed this in America. It seems the speed limits are more of a general guideline than an actual law to be followed.

  • The IIHS estimates that "[c]ompared with manual driving, the increase in speed associated with ACC/Pilot Assist use was estimated to increase crash risk by 10 percent for fatal crashes, by 4 percent for injury crashes, and by 3 percent for property-damage-only crashes." The IIHS does note that it did not take into account the following distances while using ACC, which drivers can control (in increments of 1 and 5 mph).

    That's a huge omission in their analysis. It'd also be interesting to know whether the IIHS also neglected (as in negligence) to measure or consider ACC's potentially more reliable crash avoidance features over an engaged but potentially inattentive driver.

    • Yeah... they only called out a subset of the bad outcomes. Sneaky bastards...

    • The IIHS does note that it did not take into account the following distances while using ACC, which drivers can control (in increments of 1 and 5 mph).

      That's a huge omission in their analysis.

      That's because they apparently don't even understand it at all. ACC is controllable in terms of following seconds, not mph. WTF is mph in terms of following distance; at best it could be a closing rate but that's not what one sets with these systems.

  • Read the comments at the end of the article. They pretty much explain why this is a stupid study with stupid conclusions.

  • Obviously (Score:4, Interesting)

    by drinkypoo ( 153816 ) <drink@hyperlogos.org> on Saturday March 13, 2021 @12:38AM (#61153132) Homepage Journal

    Interestingly, drivers sped more on highways with 55 mph and 60 mph speed limits than on 65 mph-limited roads.

    The faster you're going, the less difference there is between the speed you're going, and the speed you want to be going. And most people only want to go so fast before their self-preservation instinct kicks in.

    I haven't had radar cruise control yet, but I don't do a lot of driving in heavy traffic so I also don't need it. But when I use regular cruise control, I don't wind up accidentally using extra pedal because some part of me wants to go faster. I only use extra pedal when I choose to deliberately, for passing. Radar cruise control only makes you speed more if you ask it to, so people must be doing that. I'm surprised that the automakers are even allowed to make it offer you speeds in excess of about 70 mph in the US. And really, if you're going much faster than that, you probably shouldn't have cruise control on anyway.

    • Re:Obviously (Score:4, Informative)

      by ShadowRangerRIT ( 1301549 ) on Saturday March 13, 2021 @12:51AM (#61153152)

      I'm surprised that the automakers are even allowed to make it offer you speeds in excess of about 70 mph in the US.

      Highways with speed limits up to 80 MPH are common across the northern Rockies and Plains states, and in isolated regions elsewhere. I've definitely used cruise control at speeds of up to 84 MPH (my tires were only rated for 85 MPH and below) on a cross-country jaunt along I-90 through Montana-Wyoming-South Dakota. It's not that unsafe; the roads are largely empty and (especially in South Dakota) you can see any turns or other cars miles away.

      • Yeah, I frequently set my cruise to 85 or 87 on 80 MPH freeways. When surrounded by miles of emptiness and driving on road that's wide, flat, straight, dry, and clear of obstacles, there's no reason not to.
    • The highway speed limit in my city of about 250K was only recently, within the last five years, reduced to 70 mph within the city limits. It’s still 75 mph as soon as you get out of town, and people still frequently drive at 75-80 through town, just as they did for decades. I also live in one of the handful of states where exceeding the speed limit is not, in and of itself, illegal, so a speed governor of that sort would be limiting potentially legal activity.

  • I am much more likely to speed when driving myself, rather than when I have Autopilot engaged.

    • by evanh ( 627108 )

      As was noted in the article, the following distance wasn't accounted for. The closer you follow the more you impact the driver behaviour in front of you. So, it's not about your speed so much as it is about the front driver's speed.

      • by Whibla ( 210729 )

        As was noted in the article, the following distance wasn't accounted for. The closer you follow the more you impact the driver behaviour in front of you. So, it's not about your speed so much as it is about the front driver's speed.

        If someone wants to follow 'that' close I increase the follow distance I'm leaving to the car in front of me. My speed is not affected in the long term, it is still determined by prevailing traffic conditions (i.e. the car in front).

        Additionally, if the opportunity arises I'll happily get out of their way to let them overtake me (in the UK it's not free for all traffic flow, 'undertaking' is not legal).

        I have no desire to get rear ended by some muppet with delusions of being the next F1 champion.

        • in the UK it's not free for all traffic flow, 'undertaking' is not legal

          As a policeman told us during a safe driving course: in the UK, it's legal to overtake on either side on a one-way street, and what's a motorway, but two one-way streets?

          • by Whibla ( 210729 )

            in the UK it's not free for all traffic flow, 'undertaking' is not legal

            As a policeman told us during a safe driving course: in the UK, it's legal to overtake on either side on a one-way street, and what's a motorway, but two one-way streets?

            It appears I was mistaken in that. I really should look these things up before shooting my mouth off. :-/

            Thanks for the correction.

          • by Whibla ( 210729 )

            My previous post notwithstanding, the highway code states:

            "Rule 267
            Do not overtake unless you are sure it is safe and legal to do so. Overtake only on the right."

            "Rule 268
            Do not overtake on the left or move to a lane on your left to overtake."

            And, in terms of legality a quick search turned up the following: "In fact, the Highway Code wording makes it clear to identify the law. The words of MUST and MUST NOT are the rules. Disobeying national highway laws is committing a criminal offence."

            It's over 30 years

            • Read the rest of rule 268.

              • by Whibla ( 210729 )

                I'm not sure what point you're trying to make here.

                I had read it earlier, but as it specified congested traffic conditions it didn't seem relevant to the rules regarding overtaking / undertaking in free flowing traffic.

                Either way, like I said in my first reply to you, I was wrong about it being illegal to undertake. My second reply was just a clarification, pointing out the difference between the 'rules of the road' and the law, in case anyone was interested.

  • Not on longer trips on interstate highways. On longer trips you are more likely to go slower than average traffic to better balance range with charging times. Rather get there 10 minutes later from driving the speed limit or 40 minutes later because you had to stop and charge?

    Anyway, ACC just keeps you safely behind another guy if you don’t adjust it to minimal follow distance. ACC reduces stress, roadrage, and annoyances if used correctly from any manufacturer. I’ve had it on cars since about 2

  • If the car companies want to sell cars in the EU after 2023, they will have to anyway, the new cars won't be able to speed anymore, since the local speed limit will be enforced by the car's computer by law.

  • In 2009 I replaced my Honda Accord with a Volvo S80 V8.

    It wasn't quite a "sports car" but it was definitely pretty easy to drive fast and very stable/comfortable on rural highways at 90 MPH and anecdotally I knew of at least one driver who covered a lot of rural miles at 110-120 mph. So when I first got it, I drove fast a lot.

    But it had adaptive cruise control and I quickly got used to it. So used it than in normal metro area freeway driving I just quit bothering to speed. Speeding on metro freeways requ

  • >"Drivers Are 24% More Likely To Speed When Using Adaptive Cruise Control"

    So what?
    Driving somewhat over the speed limit is a non-issue and the overwhelming norm. Speed isn't the problem, it is major speed DIFFERENTIAL that is dangerous. It is also FOLLOWING DISTANCE that is dangerous. ACC actually addresses both- it reduces differential and enforces safe following distances.

    Anyway, I hate ACC- I have had it for over 12 years and never use it because, for some reason, I completely lose focus when using

    • by Shaitan ( 22585 )

      They can't even make a ticket stick at 5 over..

    • Speed isn't the problem, it is major speed DIFFERENTIAL that is dangerous.

      The faster you go, the greater the speed differential between you and stationary objects, and you say the speed differential is the danger, but you also say that speed isn't the problem. Make up your mind.

      • >"The faster you go, the greater the speed differential between you and stationary objects"

        I am not talking about stationary objects.

        >"and you say the speed differential is the danger, but you also say that speed isn't the problem. Make up your mind."

        Speed, itself, doesn't cause accidents. It can and does aggravate accidents that do occur, and those happen because of other factors. Speed differential is that between the vehicles. Some cars going 85, others going 50, for example. And that differen

  • I remember a PSA way back trying to browbeat people into obeying the speed limit by saying the driver (who just got pulled over) would save only 3 minutes. Yeah, okay, but if he does that twice a day, that's 6 minutes. If he does that every work day ((50-2)*5)=250), that works out to 25 hours every year. That works out to rather a lot of money ($1250 if you get paid $50 an hour).

    Ultimately, once this technology is standard in all cars, there will be a big boost in productivity and/or free time. Of cours

    • by saying the driver (who just got pulled over) would save only 3 minutes.

      That's over longer distances than a hop to the next town. When I used to take my parents to New York, their GPS would give us an estimated arrival time based on the speed limits between their house and Hoboken.

      I set the cruise control for five miles over the speed limit. Over the three hour drive we could see our arrival time come down by about five minutes. So yes, 3 minutes might be saved, but only if you're driving far enough.

    • If he does that every work day ((50-2)*5)=250), that works out to 25 hours every year. That works out to rather a lot of money ($1250 if you get paid $50 an hour).

      Time isn't money, though. They aren't directly interchangeable. There are always other considerations, like could and/or would you actually get paid for that time.

  • by sarren1901 ( 5415506 ) on Saturday March 13, 2021 @11:59AM (#61153976)

    When I got my Honda Insight 2019 I was super excited about the ACC feature. I messed with it for a couple of weeks before deciding it was more of a pain then I wanted to bother with.

    The biggest problem for me is when you have the car on ACC and you go around a bend on a multi-lane highway. If the bend was sharp enough, a car in the lane to your left or right will briefly be in front of you and the car will slow down. This is terrible when everyone is going about the same speed and there is zero reason to slow down at all. The person behind you has to slow down because you slowed down, literally for no reason.

    With traditional CC, you would keep going forward whatever speed you set.

    Also, with ACC, people would constantly jump in my "gap" and then the car would slow down even more.

    I quickly gave up using it. It's nice my car has some emergency braking but having your car slow down arbitrarily because the car in the lane next to you in a few feet ahead is horrible.

    • by ceoyoyo ( 59147 )

      I guess they're still working the bugs out. The one on my Mazda works quite well. It will slow down when someone jumps into the gap ahead of you, but that's what it's supposed to do. It doesn't brake abruptly though, unless the merging car is going slower.

      • I guess they're still working the bugs out. The one on my Mazda works quite well. It will slow down when someone jumps into the gap ahead of you, but that's what it's supposed to do. It doesn't brake abruptly though, unless the merging car is going slower.

        The car should immediately try to cut them off if it looks like they are going to pull in front of you. Failing that, it could purposely drive extremely close and maybe weave left and right and blow the horn at the car that cut them off.

        Ok, maybe not.

  • To test how ACC affected speeding, the IIHS recruited 40 drivers in the Boston area

    Boston drivers?!?! holup

  • "[c]ompared with manual driving, the increase in speed associated with ACC/Pilot Assist use was estimated to increase crash risk by 10 percent for fatal crashes, by 4 percent for injury crashes, and by 3 percent for property-damage-only crashes"

    It's obvious they did not gather the data for this "estiamte" from the study itself, which only had 40 drivers (unless they had multiple fatalities.) In fact the "estimation" sounds like they assumed that all types of speeding are the same in terms of risk. As in, th

  • The fact is, that police in America give you up to 10 mph over. Hit 11, sometimes 10, and they will happily ticket you.

    As to the safeness of ACC with this, I would guess that this is much safer, even doing 10-15 mph over the limit, than is a manual driver doing the speed limt. The reason is that the vehicle is far more attentive than we humans are. Our tesla slows down, speeds up, etc. in a nice fashion. Overall, I suspect that as ACC hits all of the cars, it will decrease accidents a great deal, until f
  • So... People without cruise control drive 61 on roads with a speed limit of 55 and people using adaptive cruise control drive about 62mph....

    And this difference accounts for 10% increase in fatality rate... Sure.

    (that "sure" was meant to have a longer/repeated/drawn out letter in the middle, but they won't let me post that. Sigh).

  • "It also noted that drivers generally drive 5-10 mph (8-16 km/h) faster than the posted limit, and that's easier to accomplish when using ACC than when driving manually." Why is driving faster easier with ACC? I used to set my cruise control exactly at the speed limit and cruise in the slow lane. Setting ACC to be faster than the average manual speed for the same driver is not at all obvious, so it's incumbent upon the researchers to explain this.

    It's obvious that manual driving should have a far higher

"I've seen it. It's rubbish." -- Marvin the Paranoid Android

Working...