Status Report Knowledge gap New studies highlight driver
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
Status Report |
Insurance Institute for Highway Safety Highway Loss Data Institute
Knowledge
gap New studies highlight driver
confusion about automated systems
ALSO IN Where automation is used
THIS ISSUE
Repair costs for aluminum F-150
Vol. 54, No. 4
June 20, 2019 LATCH ease-of-use ratings improve
Tesla Model 3 Autopilot
V
ehicles are getting increasingly Level 1 or 2, which applies to systems that For the survey, more than 2,000 driv-
sophisticated, with more and more can perform one or more parts of the driv- ers were asked about five Level 2 system
of them able to stay in a lane and ing task under supervision of the driver. An names currently on the market. The names
maintain a set speed and following dis- example of a Level 1 system is lane center- were Autopilot (used by Tesla), Traffic Jam
tance with minimal driver input. But this ing, in which lateral control of the vehicle Assist (Audi and Acura), Super Cruise (Ca-
kind of automation has limitations that can is automated, or adaptive cruise control, in dillac), Driving Assistant Plus (BMW) and
be tricky for drivers to grasp, and two new which longitudinal control — i.e. speed and ProPilot Assist (Nissan). Participants were
IIHS studies highlight misperceptions or following distance — is automated. Sys- told the names of the systems but not the
gaps in drivers’ understanding. tems that can perform both of those func- vehicle brands associated with them and
One study revealed how the names man- tions simultaneously are Level 2 systems. weren’t given any other information about
ufacturers use for these systems can send These systems are a far cry from Level the systems.
the wrong messages to drivers regard- 5 automation, in which the entire driving None of these systems reliably manage
ing how attentive they should be. Another task can be performed without input from lane-keeping and speed control in all situ-
found that drivers don’t always understand a human under all conditions. ations (see Status Report, Aug. 7, 2018). All
important information communicated by of them require drivers to remain attentive,
system displays. System names and all but Super Cruise warn the driver if
“Current levels of automation could po- Despite the limitations of today’s systems, hands aren’t detected on the wheel. Super
tentially improve safety,” IIHS President some of their names seem to overpromise Cruise instead uses a camera to monitor
David Harkey says. “However, unless driv- when it comes to the degree to which the the driver’s gaze and will issue a warning if
ers have a certain amount of knowledge driver can shift their attention away from the driver isn’t looking forward.
and comprehension, these new features the road. One name in particular — Auto- Each participant answered questions
also have the potential to create new risks.” pilot — signals to drivers that they can turn about two of the systems chosen at random.
The automation available in vehicles their thoughts and their eyes elsewhere, an They were asked whether particular behav-
available for purchase today is considered IIHS survey found. iors were safe while using that technology.
2 | Status Report — Vol. 54, No.4
When asked whether it would be safe to Percent of drivers who considered behaviors safe while Level 2 system is in operation
take one’s hands off the wheel while using 50%
the technology, 48 percent of people asked
about Autopilot said they thought it would
be, compared with 33 percent or fewer for
40%
the other systems. Autopilot also had sub-
■ Autopilot
stantially greater proportions of people
who thought it would be safe to look at ■ Traffic Jam Assist
scenery, read a book, talk on a cellphone or 30% ■ Super Cruise
text. Six percent thought it would be OK to ■ Driving Assistant Plus
take a nap while using Autopilot, compared ■ ProPilot Assist
with 3 percent for the other systems. 20%
At least a few Tesla owners have been
misusing Autopilot in this way, with fatal
results. 10%
In March, a Tesla driver crashed into
the side of a tractor-trailer in Florida. The
Model 3 went completely under the truck, 0%
shearing off the Model 3’s roof and killing Hands off the Talking on a Texting Watching Taking
its driver. A preliminary investigation by steering wheel cellphone a video a nap
the National Transportation Safety Board
found that Autopilot was engaged at the Eighty volunteers viewed videos record- 2018) and changing circumstances may re-
time of the crash, and the driver’s hands ed from the driver’s perspective behind the quire the driver to intervene.
were not detected on the steering wheel. wheel of the E-Class. The participants were In the study, certain key pieces of infor-
The same was true in the crash of a Tesla asked about the operating status of the mation eluded many of the participants.
Model X in California one year before (see adaptive cruise control and lane-center- While almost everyone was able to under-
Status Report, Aug. 7, 2018) and a 2016 ing features. If the features were inactive, stand when adaptive cruise control had ad-
Florida crash of a Model S that also in- the participants were asked to explain why. justed the vehicle speed or detected another
volved the side of a tractor-trailer. Half of the participants first received some vehicle ahead, most participants, regard-
“Tesla’s user manual says clearly that the training in the form of a brief orientation less of whether they received the training,
Autopilot’s steering function is a ‘hands- about the instrument cluster icons pertain- struggled to understand what was happen-
on feature,’ but that message clearly hasn’t ing to the two systems. ing when the system didn’t detect a vehicle
reached everybody,” Harkey says. “Man- Understanding these displays is impor- ahead because it was initially beyond the
ufacturers should consider what message tant because automated systems can behave range of detection.
the names of their systems send to people.” unexpectedly (see Status Report, Aug. 7, Most of the people who didn’t »
Instrument cluster information Levels of driving automation (developed by SAE International)
If a system name is one of the first ways
a driver learns about a feature, another Level 0 The human driver does everything.
source of information is the instrument Level 1 An automated system can assist the human driver in conducting
Available
cluster. Displays are important because one part of the driving task.
on vehicles
they tell a driver how a system is respond-
Level 2 An automated system can assist the driver with multiple parts of that can be
ing to situations or when a system is tempo-
the driving task. The driver must continue to monitor the driving purchased
rarily inactive. For example, a lead vehicle
environment and be actively engaged. today
may disappear from the display when that
vehicle is cresting a hill and no longer de-
Level 3 An automated system conducts all of the driving task without driver
tected by the system’s radar. Similarly, lane
engagement and monitors the driving environment, but the human
lines may disappear from the display when
driver must stand by to intervene in response to a system failure or
the lane markings on the road are no longer
request from the system to take over.
visible to the system’s cameras.
The second recent IIHS study looked at Level 4 An automated system can conduct the entire driving task without
whether drivers understand this informa- driver input but only in certain conditions (e.g., limited to 25 mph)
tion from the display of a 2017 Mercedes- or places (e.g., a city center).
Benz E-Class with the Drive Pilot system.
The E-Class display is typical of displays Level 5 An automated system can perform the entire driving task without
from other automakers. driver input under all conditions.
June 20, 2019 |3
you need to be ready to brake. Likewise,
when lane centering does not work be-
cause of a lack of lane lines, you need to
steer,” says Harkey. “If people aren’t un-
derstanding when those lapses occur,
manufacturers should find a better way of
alerting them.”
Some systems use audible alerts in those
situations in addition to visual signals. That
2017 may help, but drivers often find audible
Mercedes-Benz E-Class alerts annoying. Another option is making
the pertinent visual signal more obvious
with Drive Pilot dash display
and understandable.
Adaptive cruise control indicator Lane-centering indicator Although systems ideally should be intu-
itive, providing an orientation at the deal-
This car icon is green when a lead vehicle This steering wheel icon is green when the
ership could also help. The study showed
is detected and gray when no lead vehicle lane-centering system is actively control-
that interface-specific training helped driv-
is detected. The leading car depicted in the ling steering and gray when it is inactive
ers notice changes in lane-centering activ-
center means the same thing, but drivers due to road conditions. From the green
ity and use the correct icons to determine
can choose to have information about other state, it sometimes briefly flashes yellow
system status.
vehicle systems displayed there instead. when transitioning from active to inactive.
For copies of “What’s in a name? Drivers’
perceptions of the use of five SAE Level 2
driving automation systems,” by E.R. Teoh
(« from p. 3) receive training also struggled participants often couldn’t explain why the and “Effects of training and interface con-
to identify when lane centering was inac- system was temporarily inactive. tent on Level 2 driving automation inter-
tive. In the training group, many people “If your Level 2 system fails to detect a face usability,” by A.S. Mueller et al., email
got that right. However, even in that group, vehicle ahead because of a hill or curve, StatusReport@iihs.org. n
Is automation used where it’s intended?
The automated systems available in vehicles today other freeway miles of Evoque drivers were higher speed limits and free-flowing traffic, and
are designed to be used only on certain types of driven using adaptive cruise control. Before the so in that sense they fit the usual criteria for
roads. A recent study by researchers from IIHS and software update, 11 percent of interstate/free- Level 1 or 2 systems. However, they often have
the Massachusetts Institute of Technology’s AgeLab way miles in the S90 were driven with adap- intersections, and many manuals advise against
found that, for the most part, drivers use the tech- tive cruise control alone and another 11 percent using the systems on roads with intersections.
nology where it was intended, though they may not were driven with the Level 2 Pilot Assist system. That’s because the systems don’t react to traffic
be using it enough to have a measurable effect on After the update, those numbers were 8 percent lights or stop signs and can have trouble detect-
safety, and some individuals may be using the sys- and 20 percent, respectively. ing stopped vehicles ahead and avoiding cross
tems on roads they weren’t designed for. “Driving automation could reduce crashes traffic. They also can have trouble staying within
For the study, a project of the Advanced Ve- by eliminating some of the potential for human the lane through intersections.
hicle Technology Consortium, volunteer drivers error,” says IIHS Senior Research Scientist Ian The Evoque’s adaptive cruise control was on for
spent four weeks driving either a Range Rover Reagan. “But given the low use of the systems 7 percent of nonfreeway principal arterial miles.
Evoque with adaptive cruise control or a Volvo and the fact that most vehicles on the road today Eleven percent of the post-update S90 nonfree-
S90 with adaptive cruise control and Pilot Assist, still don’t have these features, we don’t expect to way principal arterial miles involved Pilot Assist.
a Level 2 system that enhances the adaptive see these crash reductions any time soon.” There was wide variation in use of the systems,
cruise control with lane centering. During the Far smaller percentages of miles traveled with some drivers not using them at all and some
study, the S90 underwent a software update that on nonfreeway roads in the study involved using them a lot. One Evoque driver drove 41 per-
improved the lane centering, so the researchers automation. cent of nonfreeway principal arterial miles with
looked separately at those who drove the S90 Vehicle manuals often give ambiguous instruc- adaptive cruise control, and one post-update S90
before the update and those who drove it after. tions about where to use these types of systems, driver used Pilot Assist for 30 percent.
Like most driving automation in currently sometimes saying only that they are for “high- For a copy of “Measuring adult drivers’ use
available vehicles, these systems are meant to way use.” It’s not always clear if nonfreeway ar- of Level 1 and 2 driving automation by road-
be used on interstates and other freeways. terials should be considered “highways” or not. way functional class” by I.J. Reagan, email
In the study, 40 percent of interstate and On the one hand, these roads often have StatusReport@iihs.org. n
4 | Status Report — Vol. 54, No.4
Ford F-150 repair costs remain
steady despite use of aluminum
F
ord’s switch to aluminum for the body
of the F-150 pickup hasn’t resulted in
higher repair costs, in part because
the company is pricing the aluminum parts
lower than steel ones, HLDI analysts have
found. However, while the cost for repairs
hasn’t risen, the time required for them ap-
pears to have gone up, and that could lead
to higher insurance costs.
When Ford began building its iconic
truck out of aluminum instead of steel,
consumers had questions about the effect
it would have on safety and their wallets.
At the time, IIHS high-speed crash tests
showed safety wasn’t adversely affected,
while a separate experiment showed repair
costs were. Damage to an aluminum-body
F-150 from a low-speed crash turned out
to be more expensive to fix than damage
to an older, steel-body F-150 put through
the same crash (see “Pricier repairs for alu- Ford F-150
minum F-150 than steel model in fender-
benders,” July 30, 2015). 2014 model, while other 2016 models had However, despite the findings on claim
Now HLDI has an update based on re- claim severities 12-21 percent above their severity, HLDI uncovered some evidence
al-world claims data. In the four years 2014 model year results. of other, hidden costs. It takes longer for
since Ford introduced the aluminum-body These rising claim severities are consis- loss information to accumulate for F-150
truck, the change in material hasn’t result- tent with data HLDI has collected on claims claims, and that is likely an indication
ed in more costly insurance claims. That’s across all vehicles, which show that the av- that repairs are taking longer. That means
likely a result of Ford’s efforts to hold down erage loss payment per claim has gone up F-150 owners have to make do for longer
the price of aluminum replacement parts in recent years. Given that the F-150’s claim without their vehicles, and there are addi-
and simplify repairs. At the same time, severity didn’t rise as much as the severities tional costs to insurers for rental vehicles.
Ford has raised the prices of steel parts for for other pickups, the 4 percent increase for Although these delays decreased in the
older models. the 2016 model year is likely unrelated to 2016 and 2017 model years, repairs are still
HLDI analysts compared average loss the use of aluminum. taking longer for the aluminum truck than
payment per claim, or claim severity, under This result contrasts with an earlier for the steel F-150 and for other models.
collision insurance, which covers damage HLDI study that looked at aluminum use In addition, HLDI found that collision
to an at-fault driver’s vehicle, for the alu- in luxury vehicles and found it was asso- claim frequency is 7 percent higher for the
minum-body 2015-16 F-150 and the steel- ciated with higher collision claim severity. aluminum F-150 compared with control ve-
body 2014 model. They also compared When Ford switched to aluminum for the hicles. It’s not clear what has caused that in-
2015 and 2016 models of the Chevrolet Sil- F-150’s body in a bid to shed pounds and crease, but the aluminum-body truck may
verado 1500, GMC Sierra 1500 and Ram improve fuel economy, it took measures to be more easily damaged than the steel one.
1500 with their 2014 counterparts. Those ensure repair costs wouldn’t rise. To help “Ford has worked to keep repair costs
pickups have kept their steel bodies. make repairs simpler with lower labor costs, down for its aluminum-body truck,” says
Claim severity for each of the 2015 and the company used a modular design. Ford HLDI Senior Vice President Matt Moore.
2016 models was higher than for the cor- also offered dealerships and body shops the “But unless other manufacturers take the
responding 2014 models, but the increase chance to buy tools for repairing alumi- same steps, there’s no guarantee that these
was smallest for the F-150. It rose less than num-body vehicles at a discount. Finally, results will hold true for future aluminum-
1 percent for 2015 models, compared with it has lowered prices on many parts for the body vehicles from other manufacturers.
5-7 percent for the other pickups. For the 2015-16 F-150 while raising prices for the In addition, the higher claim rates are con-
2016 F-150, severity rose 4 percent over the same parts for the 2014 steel model. cerning.” n
June 20, 2019 |5
Vehicle manufacturers make
strides on LATCH ease of use
N
early three-quarters of 2019 vehicles — the place where the seatback meets the the seatback, under the seat, on the ceil-
have LATCH hardware that rates bottom seat cushion — or slightly deeper ing or on the floor.
good or acceptable for ease of use, if there is open access around them. 4The area where the tether anchor is
as automakers continue making improve- 4The lower anchors are easy to maneuver found doesn’t have any other hardware
ments that help parents and caregivers around. This is defined as having a clear- that could be confused for the tether
properly install child restraints. ance angle greater than 54 degrees. anchor. If other hardware is present, then
The results mark a shift from 2015, when 4The force required to attach a standard- the tether anchor must have a clear label
IIHS launched its LATCH ease-of-use rat- ized tool representing a child seat con- located within 3 inches of it.
ings. At that time, a majority of new vehi- nector to the lower anchors is less than To earn a good rating, two LATCH posi-
cles rated poor or marginal. 40 pounds. tions in the second row must meet all five
Today, 23 vehicles earn the top rating of 4Tether anchors are on the vehicle’s rear criteria, and a third tether anchor must
good+, 31 are rated good, and 87 rate ac- deck or in the middle of the seatback. meet both tether criteria.
ceptable. Forty-nine vehicles are marginal, They shouldn’t be at the very bottom of The good+ rating is for vehicles that
and only four earn a poor rating. Among
automakers, Toyota and Subaru are stand-
outs for LATCH ease of use, while U.S. au-
tomakers lag behind. Installation in pickups
is trickier than in other types of vehicles.
A properly installed, age-appropriate
child restraint can protect a child much
better in a crash than a seat belt alone.
LATCH, which stands for Lower Anchors
and Tethers for Children, is intended to
make child restraint installation easier.
Child restraints installed with LATCH are
more likely to be put in correctly than re-
straints installed using the vehicle seat belt,
IIHS research has shown.
But even with LATCH, installation isn’t
always simple, and errors are common. The
Institute’s ratings are based on ease-of-use
criteria that have been shown to minimize
mistakes.
Toyota and Subaru stand out for LATCH
ease of use. They’re tied for the most ve-
hicles with the highest rating of good+.
“With child restraints, a good, tight in-
stallation is critical but can be difficult to
achieve,” says Jessica Jermakian, an IIHS
senior research engineer. “Thanks to these
recent improvements in vehicle LATCH
hardware, we expect more children will be
riding in correctly installed seats.”
In the IIHS ratings system, LATCH hard-
ware is considered good if it meets the fol-
lowing criteria:
4The lower anchors are no more than
¾ inch deep within the seat bight Subaru Crosstrek Hybrid
6 | Status Report — Vol. 54, No.42019 vehicles with good+ and good LATCH ratings
Listed rating is the highest available for the most popular seat covering within the vehicle class.
Good+ Good
Acura RDX Subaru Crosstrek Audi A4 Hyundai Nexo Nissan Kicks
Audi Q7 Subaru Forester Audi A4 Allroad Lexus ES Nissan Maxima
Honda Accord Subaru Impreza sedan Audi A5 Coupe Lexus IS Nissan Rogue
Honda Insight Subaru Impreza wagon Audi A5 Sportback Lexus NX Nissan Sentra
Honda Odyssey Subaru Legacy Audi A6 Lexus RC Toyota C-HR
Jeep Cherokee Subaru Outback Audi e-tron Mercedes-Benz C-Class Toyota Highlander
Lexus RX Toyota Avalon Audi Q5 Mercedes-Benz E-Class Toyota Prius Prime
Lexus UX Toyota Camry Audi Q8 Mercedes-Benz GLS-Class Volkswagen GTI
Mazda 3 hatchback Toyota Corolla hatchback BMW 2 series Mercedes-Benz GLE-Class Volkswagen Passat
Mazda 3 sedan Toyota Prius BMW 3 series Mitsubishi Mirage
Mazda 6 Toyota RAV4 BMW X5 Nissan Altima
Subaru Ascent
Pickups like this one (shown with head restraints removed) typically require child seat tethers to be fed through a loop at the top of the vehicle seat (left)
and then attached to a loop or anchor above an adjacent seating position. This complexity makes it hard for pickups to earn good LATCH ratings.
meet the criteria for a good rating and pro- borrowing) and tether anchors in all rear — has a good+ designation.
vide additional LATCH-equipped seat- seating positions. The additional tether No pickups earn a rating higher than ac-
ing positions. For a two-row vehicle, that anchors must meet at least one of the two ceptable, and 14 out of 20 are rated margin-
means having a third good or acceptable tether anchor criteria. If the vehicle has a al. The main problem is the tether anchors.
LATCH seating position. The third posi- second-row center seating position, it must Because the rear seat of a pickup is right up
tion may use either dedicated anchors or have good or acceptable LATCH there against the back wall of the cab, there aren’t
anchors borrowed from other positions. In (with or without borrowing). many options for where to locate them. In
many vehicles that have lower anchors in Of all the manufacturers, Subaru and most pickups, the tether must be routed
the second-row outboard seating positions, Toyota are tied for the most good+ rat- through a loop near the head restraint and
LATCH can be used in the center position ings. Seven of Subaru’s eight vehicles earn then attached to another loop or anchor,
by “borrowing” one anchor from each side. the designation. Of the 25 rated vehicles typically in an adjacent seating position.
Some vehicles have one dedicated anchor from Toyota and its luxury Lexus brand, “When we’ve done studies observing
for the center seat and rely on a borrowed seven earn a good+ rating and another people installing child restraints, we’ve seen
anchor for the other side. seven earn a good rating. Neither Ford nor that the tether anchors in pickups are a real
For a three-row vehicle to earn a good+ General Motors have a single model with point of confusion,” Jermakian says. “We’re
rating, it must have one additional good a good or good+ rating. In Fiat Chrysler’s continuing to work with manufacturers to
or acceptable LATCH position (without lineup, one vehicle — the Jeep Cherokee come up with solutions to this issue.” n
June 20, 2019 |7Insurance Institute for Highway Safety
Highway Loss Data Institute
Status Report
New studies highlight driver confusion IIHS is an independent, nonprofit scientific and educational organization dedicated to reducing the losses — deaths, injuries and
property damage — from motor vehicle crashes.
about automated systems42
HLDI shares and supports this mission through scientific studies of insurance data representing the human and economic losses
Is automation used where it’s intended?44 resulting from the ownership and operation of different types of vehicles and by publishing insurance loss results by vehicle make
and model.
Ford F-150 repair costs remain
steady despite use of aluminum45 Both organizations are wholly supported by the following auto insurers and funding associations:
Vehicle manufacturers make MEMBER GROUPS Liberty Mutual Insurance
strides on LATCH ease of use 46 AAA Carolinas Louisiana Farm Bureau Mutual Insurance Company
Acceptance Insurance The Main Street America Group
Alfa Insurance MAPFRE Insurance Group
Allstate Insurance Group Mercury Insurance Group
Vol. 54, No. 4 American Agricultural Insurance Company MetLife
American Family Insurance Mississippi Farm Bureau Casualty Insurance Company
June 20, 2019 American National MMG Insurance
Ameriprise Auto & Home Munich Reinsurance America, Inc.
Amica Mutual Insurance Company Mutual Benefit Group®
Auto Club Enterprises Mutual of Enumclaw Insurance Company
Auto Club Group Nationwide
Auto-Owners Insurance NJM Insurance Group
BITCO Insurance Companies Nodak Insurance Company
California Casualty The Norfolk & Dedham Group®
Inquiries/print subscriptions: Celina Insurance Group North Carolina Farm Bureau Mutual Insurance Company
Central States Health & Life Co. of Omaha and Affiliates Northern Neck Insurance Company
StatusReport@iihs.org CHUBB NYCM Insurance
Colorado Farm Bureau Mutual Insurance Company Ohio Mutual Insurance Group
Commonwealth Casualty Company Oregon Mutual Insurance Company
Copy may be republished with attribution.
Concord Group Insurance Companies Pekin Insurance
Images require permission to use. COUNTRY Financial PEMCO Insurance
CSAA Insurance Group Plymouth Rock Assurance
Desjardins Insurance Progressive Insurance
Editor: Sarah Karush DTRIC Insurance PURE Insurance
Art Director: Steve Ewens ECM Insurance Group Qualitas Insurance Company
Elephant Insurance Company Redpoint County Mutual Insurance Company
Photographers: Steve Ewens, EMC Insurance Group The Responsive Auto Insurance Company
Craig Garrett, Dan Purdy Erie Insurance Group Rider Insurance
Esurance Rockingham Insurance
Farm Bureau Financial Services RSA Canada
Farm Bureau Insurance Company of Michigan Safe Auto Insurance Company
Farm Bureau Insurance of Tennessee Safeco Insurance
Farm Bureau Mutual Insurance Company of Idaho Samsung Fire & Marine Insurance Company
Farmers Insurance Group SECURA Insurance
Farmers Mutual of Nebraska Selective Insurance Company of America
Florida Farm Bureau Insurance Companies Sentry Insurance
/iihs.org Frankenmuth Insurance Shelter Insurance®
Gainsco Insurance Sompo International
GEICO Corporation South Carolina Farm Bureau Mutual Insurance Company®
@IIHS_autosafety The General Insurance Southern Farm Bureau Casualty Insurance Company
Georgia Farm Bureau Mutual Insurance Company State Farm Insurance Companies
Goodville Mutual Casualty Company Stillwater Insurance Group
@iihs_autosafety Grange Insurance Swiss Reinsurance Company Ltd
Grinnell Mutual Texas Farm Bureau Insurance
Hallmark Financial Services, Inc. The Travelers Companies, Inc.
IIHS
The Hanover Insurance Group United Educators
The Hartford USAA
iihs.org/rss Haulers Insurance Company, Inc. Utica National Insurance Group
Horace Mann Insurance Companies Virginia Farm Bureau Mutual Insurance
Imperial Fire & Casualty Insurance Company West Bend Mutual Insurance Company
iihs.org Indiana Farm Bureau Insurance Western National Insurance Group
Indiana Farmers Insurance Westfield
Infinity Property & Casualty
Kemper Corporation FUNDING ASSOCIATIONS
Kentucky Farm Bureau Mutual Insurance Companies American Property Casualty Insurance Association
La Capitale General Insurance National Association of Mutual Insurance CompaniesYou can also read
