vehicle safety initiatives at iihs: crashworthiness, … vehicle safety initiatives at iihs:...

iihs.org

Vehicle Safety Initiatives at IIHS:Crashworthiness, Crash Avoidance,and Automated Driving

Adrian LundPresident, IIHS and HLDI

Automotive Safety Council Annual General MeetingAventura, FL

March 24, 2017

Motor vehicle crash deaths have declined dramatically

during the past 50+ years, but increased in 2015

30,000

35,000

40,000

45,000

50,000

55,000

60,000

1950 1960 1970 1980 1990 2000 2010 20150

10

20

30

40

50

60

70

80

Motor vehiclecrash deaths

Crash deaths perbillion vehicle miles traveled

201535,092 deaths

11.2 per billion

U.S. motor vehicle crash deaths and deaths

per billion vehicle miles traveled

1950-2015

Percent U.S. drivers using cellphones at any given daylight time and motor vehicle crash deaths2000-15

0

10,000

20,000

30,000

40,000

50,000

0

4

8

12

16

2000 02 04 2005 06 07 08 09 2010 11 12 13 14 2015

estimated hand-held and hands-free phone conversation

observed hand-held phone conversation

observed manipulation of hand-held devices

annual deaths

U.S. motor vehicle crash deaths and unemployment rate1950-2015

0

5

10

15

20

30,000

35,000

40,000

45,000

50,000

55,000

60,000

1950 55 60 65 70 75 80 85 90 95 2000 05 10 15

Motor vehicle crash deaths

Unemployment rate

5 percent

35,092

Crashworthiness

Front small overlapBegan in 2012

IIHS crashworthiness tests

Front moderate overlapBegan in 1995

Side impactBegan in 2003

Rear crash (whiplash mitigation)

Began in 2004

Roof strengthBegan in 2009

Crash protection ratings by model yearImprovements beginning in 1995

0%

20%

40%

60%

80%

100%

1995 1997 1999 2001 2003 2005 2007 2009 2011 2013 2015

0%

20%

40%

60%

80%

100%

1995 1997 1999 2001 2003 2005 2007 2009 2011 2013 2015

0%

20%

40%

60%

80%

100%

1995 1997 1999 2001 2003 2005 2007 2009 2011 2013 2015

0%

20%

40%

60%

80%

100%

1995 1997 1999 2001 2003 2005 2007 2009 2011 2013 2015

0%

20%

40%

60%

80%

100%

1995 1997 1999 2001 2003 2005 2007 2009 2011 2013 2015

moderate overlap front

roof strength small overlap front

side impact head restraints and seats

poor

marginal

acceptable

good

Death and injury reductionsGood versus poor in IIHS tests

Front offset with moderate overlap test

– Fatality risk in head-on crashes is 46 percent lower

Side impact crash test

– Fatality risk in side impact crashes 70 percent lower

– In addition to the benefit of adding side airbag protection for the head

Rear impact test (seat only)

– Neck injury risk in rear crashes is 15 percent lower

– Risk of neck injury requiring 3+ months treatment is 35 percent lower

Passenger side small overlap protection

40 mph small overlap front ratingsBy model year, as of February 2017

Good

Acceptable

Marginal

Poor

Small overlap countermeasures are not always applied to the passenger-side

Passenger-side moderate overlap testsof two vehicles

Moderate overlap tests indicate good protection for the passenger and performance is not affected by small overlap countermeasures

Many vehicles sold and tested asright-hand drive in other markets

Small overlap structure on driver-side

Toyota RAV-4 has driver-side only countermeasures

Rating comparisonToyota RAV-4

structure

injury

head/neck

chest

thigh/hip

leg/foot

restraints and kinematics

Good Poor

driverright front

passengerdriver

How do Good rated vehicles perform on the passenger-side?Small SUVs with variety of countermeasures

Buick Encore Mazda CX-5

Hyundai Tucson Subaru Forester Toyota RAV-4

Honda CR-V Nissan Rogue

Driver/passenger small overlap crash ratings

driver-side

impact

passenger-side

impact

visible design

application

2016 Hyundai Tucson symmetric

2015 Buick Encore symmetric

2015 Honda CRV symmetric

2015 Mazda CX-5 symmetric

2014 Subaru Forester symmetric

2015 Nissan Rogue driver-side

2015 Toyota RAV4 driver-side

Next steps for passenger-side small overlap crashes

Testing of passenger side small overlap protection will begin this year (2017)

– 1st vehicle group will be midsize cars (moderately priced)

2018 Top Safety Pick+ will require at least acceptable rating for passenger side small overlap protection

A protocol has been proposed for rating passenger side performance (available on IIHS website)

Automakers with Good performance on driver side small overlap protection can submit their test results, following usual IIHS rating verification procedures, to qualify for TSP+ if their vehicle has not been tested by IIHS

Crash avoidance testing for consumer information:

Headlights &front crash prevention

Front crash prevention systems arereducing police-reported rear-end strikes

Compared with vehicles without any front crash prevention…

…vehicles with forward collision warning only are

27% less likely to rear-end another vehicle.

…vehicles with forward collision warning AND autobrake are

50% less likely to rear-end another vehicle.

If every vehicle on the road had forward collision warning with

autobrake in 2014, there would have been an estimated

1,000,000 fewer police-reported crashes

400,000 fewer police-reported injuries

Front crash prevention ratingsBegan in 2013 after research showing that forward collision warning

and automatic braking systems are reducing crashes

Rating scalefor systems offered as standard or optional

Front crash prevention ratings2013-17 models, as of February 2017

141

40

11 10

115

53

28

19

85

62

44

34

58 59

47

61

40

58

38

82

0

40

80

120

160

2013

2014

2015

2016

2017

20 automakers have committedto make AEB a standard feature

by September 2022

99+% ofU.S. market

Motivation for headlight evaluation program

16,768 annual crash deaths during dusk/dark/dawn hours

Research has established link between obstacle detection

and improved lighting

Federal standard produces wide range of on-road visibility

Large variation in illumination

Performance is not measured when installed, so factors like lamp height and spread are not captured

Aim is not regulated

Vehicle approaches:

– 500 ft. radius left and right curves at 40 mph

– 800 ft. radius left and right curves at 50 mph

– Straightaway at 40 mph

Record illuminance readings for:

– Visibility – edges of road at 10 in. above ground

– Glare – center of oncoming lane (3 ft. 7 in.)

Dynamic headlight test setup

800 ft. radius

500 ft. radius

straightaway

direction of travel

Light sensor array

Headlight releases

Small SUV ratings

July 2016

21 models evaluated

47 headlight combinations

Midsize ratings

March 2016

31 models evaluated


Pickup truck ratings

October 2016

11 models evaluated


Consumer comments on headlight ratings

I wanted to thank IIHS for the headlight ratings report that you

released last week.

-EH (Medford, New Jersey)

I own a 2013 Ford Edge. It should have come with a Seeing Eye Dog.

For the first time in my life, I am afraid to drive at night.

-AM (Buckingham, Virginia)

Thank you for proving to my friends that I’m not crazy or blind.

-RW (Mentor, Ohio)

Thanks for the great work!

-RV (Tiverton, Rhode Island)

BMW 3 series halogen and Toyota Prius v LED Low beams

100 ft 200 ft 300 ft 400 ft

Toyota Prius v LED and BMW 3 series halogen On-road comparison

GOOD11 vehicles

ACCEPTABLE50 vehicles

MARGINAL44 vehicles

POOR91 vehicles

2017 headlightratingsAll trims andpackages testedas of February 2017

Next steps for headlight testing

Protocol was modified in late 2016 to provide a variable demerit for glare (as done with lack of illumination)

– All evaluations reflect the modification

Midsize SUVs are being evaluated now

TOP SAFETY PICK+ 2017 already requires at least an acceptable rating for headlamps

TOP SAFETY PICK+ 2018 will require a good rating and headlamps must be at least acceptable for TOP SAFETY PICK

2016-2018 TOP SAFETY PICK awards

Driver side Small

overlap front

Passenger side small

overlap front

Moderate overlap

frontSide Roof Rear

FrontCrash

Prevention

Head-lamps

2016 TSPNot

requiredBasic n/a

2016 TSP+Not

requiredAdvanced

or Superiorn/a

2017 TSPNot

requiredAdvanced

or SuperiorNot

required

2017 TSP+Not

requiredAdvanced

or SuperiorGood or

Acceptable

2018 TSPNot

requiredAdvanced

or SuperiorGood or

Acceptable

2018 TSP+Good or

AcceptableAdvanced

or SuperiorGood

G

G

G

G

G

G

G

G

G

G

G

G

G

G

G

G

G

G

G

G

G

G

G

G

G

G

G

G

G

G

Rear autobrake

Annual estimates (NHTSA)

– 267 deaths

– 15,000 injuries

2002-03 IIHS drive-in claims study (509 claims)

– 76 (15 percent) claims involved backing

5 makes currently offer rear autobrake

– Cadillac

– Chrysler/Jeep

– Infiniti

– Subaru

– Toyota

Backing crashes

IIHS/RCAR autobrake procedure(RCAR)

IIHS has been working with RCAR to develop a rear autobrake procedure

RCAR is a global association of insurance research centers dedicated to improving vehicle safety, damageability, repairabilityand security

Includes 24 centers in 19 countries on 5 continents

Working groups include Damageability, Primary Safety (PSAFE), Repairability, etc.

Jeep CherokeeCar to car

Results: car-to-car

0.4 m overlap 45° 10°

Vehicle Direction

Short Range Long Range

overlap 45° 10° overlap 45° 10°

Jeep Cherokee

Straight 2/2 2/2 2/2 2/2 2/2 2/2

Left 2/2 2/2 2/2 2/2 2/2 2/2

Right 2/2 2/2 2/2 2/3 0/2 2/2

Subaru Forester

Straight 2/2 2/2 2/2 2/2

Left 2/2 2/2 2/2 2/2

Right 2/2 0/2 2/2 0/2

Cadillac XT5

Straight 2/2 2/2 2/2 0/2

Left 2/3 2/3 2/2 0/2

Right 2/2 2/2 2/2 2/3

Infiniti Q50

Straight 0/2 2/2 1/2 0/2 2/2

Left 2/2 2/2 2/2 2/2 2/2

Right 2/2 2/2 2/2 1/3 2/2

Cadillac CTS

Straight 2/2 2/3 2/2 2/2 0/2

Left 2/2 2/3 2/2 1/3 2/2

Right 2/2 2/2 2/2 0/2 0/2

All avoidance All impactsSome impacts/some avoidance

Cadillac CTSCar to pole

Results: car-to-bollard and car-to-pillar

Center Bollard 0.4 m Bollard Center Pillar 0.4 m Pillar


Vehicle Distance

Test Scenario

Center

Bollard

0.4m

Bollard

Center

Pillar

0.4 m

Pillar

Subaru ForesterShort Range 2/2 2/2 2/3 2/2

Long Range 2/2 2/2 2/2 0/2

Cadillac CTSShort Range 2/2 2/2 2/2 2/2

Long Range 2/3 2/2 0/2 0/2

Cadillac XT5Short Range 0/2 2/2 0/2 2/3

Long Range 2/2 2/2 0/2 2/2

Jeep CherokeeShort Range 2/3 2/2 0/2 0/2

Long Range 0/2 0/2 0/2 0/2

Infiniti Q50Short Range 0/2 0/2 0/2 0/2

Long Range 2/2 0/2 0/2 0/2

NHTSA rear autobrake procedureIssued as part of NCAP proposal

Vehicle starts 20 feet from a stationary child pedestrian mannequin

– 6 year-old, roughly 3 feet 10 inches tall

Shift vehicle into reverse and coast rearward

Tests conducted with pedestrian in 3 different locations

– Centerline

– 2 feet toward the vehicle’s passenger side

– 2 feet toward the vehicle’s driver side

Results: NHTSA procedure


Vehicle

Test Scenario

2 feet left Center 2 feet right

Cadillac CTS 2/2 2/2 2/2

Cadillac XT5 2/2 2/2 2/2

Subaru Forester 2/2 2/2 2/2

Jeep Cherokee 0/2 0/2 2/3

Infiniti Q50 0/2 0/2 0/2

Automated driving

“How driverless cars will radically change every aspect of our lives”

“Google’s self-driving cars have

autonomously driven over 1 million miles”

“Honda says autonomous cars won’t be ready until 2030 at the earliest”

“Nissan aims for fully autonomous cars by 2020”

Are autonomous vehicles the “next big thing?”Recent headlines

Driver acceptance of systems

Percent of vehicle owners who reported driving with forward collision warning turned on

0

20

40

60

80

100

Toyota Volvo Dodge and Jeep

unknown

never

sometimes

always

On-off status of front crash prevention systemsBy manufacturer

percent with

system on

number

observed

Cadillac 92 206

Chevrolet 87 142

Honda 98 239

Lexus 50 8

Mazda 95 20

Volvo 94 52

total 93 667

Percent of vehicle owners who reported driving with lane-maintenance systems turned on

0

20

40

60

80

100

Volvolane departure warning

Infinitilane departure warning

Infinitilane departure prevention

Toyotalane departure prevention

unknown

never

sometimes

always

On-off status of lane-maintenance systemsBy manufacturer

percent with

system on

number

observed

Cadillac 57 207

Chevrolet 50 147

Ford/Lincoln 21 115

Honda 36 239

Lexus/Toyota 68 147

Mazda 77 26

Volvo 75 105

total 51 986

On-off status by maximum observablelane-maintenance intervention levelPercent with system on

0

20

40

60

80

lane departure warning(n=547)

lane departure prevention(n=288)

active lane keeping(n=148)

Experiences with vehicle technology

Driver experience program

Driver assistance and driving automation systems change the driving task and the driver’s role

Level 2 driving automation requires the driver to remain engaged in the driving task

How do drivers interact with technologies that assist with or automate parts of the driving task?

Vehicles

2016 Toyota Prius2016 Infiniti QX60 2016 Honda Civic

2017 Audi Q7 2017 Audi A4

Recorded information from over 60,000 miles and 2 years of daily driving

phase 1 phase 2

March - July 2016 August 2016 - January 2017

employee drivers 54 47

vehicle uses 80 80

reported miles driven 33,584 31,331

reported days of driving 354 423

First phase focused on driver interactions with various technologiesPost-use survey topics

Participants indicated their level of agreement with various statements about:

– Trust

e.g., system is dependable, suspicious of system, system is reliable

– Ease of use

e.g., easy to use, functions well-integrated, learned to use quickly, confident in using

– Comprehension of system displays and status

e.g., information located where expected, easy to understand setting or status

Reported likes, dislikes and unexpected system behavior

Side-view assist ranked first in trustAverage rating and 95% confidence interval by system

side-view assist

(Honda, Audi, Infiniti)

lane departure warning

forward collision warning

active lane keeping

(Honda, Audi)

adaptive cruise control

strongly

disagree

disagree neutral agree strongly

agree

Honda’s ACC system scored lowest in trustAverage rating and 95% confidence interval by vehicle

Honda Civic

Audi Q7

Toyota Prius

Infiniti QX60

strongly

disagree


agree

Infiniti’s side-view assist was trusted the leastAverage rating and 95% confidence interval by vehicle

Honda Civic

Audi Q7

Infiniti QX60

strongly

disagree


agree

Technologies had different problem areasPercentage of drivers by complaint type

0

20

40

60

80

adaptive cruisecontrol

active lane keeping lane departurewarning

forward collisionwarning

side view assist

functionality and performance user interface circumstance none


0

20

40

60

80




side view assist

“…[ACC] approaches too close for comfort when cars in

front slow down quickly.”

Participant 302CV

“[ACC] didn't decelerate fast enough for stopped traffic at

the bottom of hills.”

Participant 109SA



0

20

40

60

80




side view assist

“You have to get in a wrestling match for control of

the wheel on curves.”

Participant 121SA

“I did not feel well notified by the system of when it

was on and had been working, but no longer could

identify the lines.”

Participant 329SV

“On highways this worked well, but … the constant

pressure in one direction was fatiguing.”

Participant 301SV


Drivers complained about Honda’s ACC system performancePercentage of driver comments about ACC by complaint type

0

20

40

60

80

100

Honda Civic Infiniti QX60 Toyota Prius Audi Q7


Drivers complained about Honda’s ACC system performancePercentage of driver comments about ACC by complaint type

0

20

40

60

80

100


“Lags and sudden acceleration/deceleration in more

congested situations.”

Participant 101CA

“It also seemed to brake very abruptly and hard.”

Participant 116SA

“It did not gradually slow down, it would brake too hard

when not necessary and then accelerate.”

Participant 326SV


Drivers also complained about Audi’s ACC system performancePercentage of driver comments about ACC by complaint type

0

20

40

60

80

100


functionality and performance user interface circumstance none“The acceleration response when changing lanes to

around a slower vehicle.”

Participant 301CV

“…integration with Nav system where vehicle would

slow automatically for turns…”

Participant 327SV

“The predictive function that adjusted for school

zones, etc.”

Participant 314SV

“Would lower speed for school zones that were not

active”

Participant 320SV

“…it would reduce speed very quickly which made

me very nervous of someone hitting me...”

Participant 321SV

Second phase focused on collecting information about using automation in specific situations

Overall, I felt this technology improved my driving experiencePercentage of drivers who agreed or strongly agreed, by technology

0%

20%

40%

60%

80%

100%

2016HondaCivic

2016InfinitiQX60

2016ToyotaPrius

2017AudiA4

2017AudiQ7

2016HondaCivic

2017AudiA4

2017AudiQ7

adaptive cruise control active lane keeping

I feel comfortable using adaptive cruise control when traveling on…Percentage of drivers who agreed or strongly agreed

0%

20%

40%

60%

80%

100%

free-flowinginterstates

major arterialswith signalizedintersections

roads withmoderate hills

stop-and-go traffic low-speed,local roads

I feel comfortable using active lane keeping when traveling on…Percentage of drivers who agreed or strongly agreed

0%

20%

40%

60%

80%

100%

free-flowinginterstates

interstates withgentle to

moderate curves

roads withmoderate hills

winding, curvyroads

Tesla monitored driving(Auto-Pilot 7.1, January 2016)

Tesla driving“Autopilot” version 7.1 released January 2016

“Autopilot” features

– Traffic-aware cruise control

– Autosteer

– Autopark: added Summon feature and perpendicular parking

Enhanced driver information in instrument panel

– Can display multiple vehicles in front of Model S including cars, trucks and motorcycles, and curved lines

Added safety restriction

– Autosteer is restricted on residential roads without a center divider; speeds are limited to the speed limit plus 5 mph

Tesla “Autopilot” – The GoodDriver information

Tesla “Autopilot” – The GoodAutosteer shutdown

Tesla “Autopilot” – The BadLost lane lines

Tesla “Autopilot” – The Ugly

SAE International’s automation levels

Who or what is driving?

Fallback

Where andwhen does it

operate?Sustained

controlDetection &response

Level 0: none None N/A

Level 1: assistance Limited

Level 2: partial Limited

Level 3: conditional Limited

Level 4: high Limited

Level 5: full Unlimited

+

Initial thoughts on automated driving

The acceptance of automated driving, as with ADAS, will vary among drivers

– Benefits of automated driving overestimated in near term

Drivers may not distinguish among levels of autonomy

– As level 2 systems proliferate and become more dependable, they will be treated as level 3 or 4

– Restrict level 2 systems to their intended operational domains

The system should not surprise the driver

– Disengagements should be clear

– Inadvertent driver disengagement should be difficult

– A vehicle should not wander ”hunting” for missing lines

System disengagement should begin to slow the vehicle until driver demonstrates control

USA Today

December 29, 2016

iihs.org

More information and links to our YouTube channeland Twitter feed at iihs.org

Adrian LundPresident, IIHS and [email protected]

vehicle safety initiatives at iihs: crashworthiness, … vehicle safety initiatives at iihs:...

Documents