Transcript Document

On-Board Technologies Applicable
to Teen Drivers:
A Review
Shawn Brovold, Max Donath, Craig Shankwitz, Nic Ward
ITS Institute, University of Minnesota
(Contact: [email protected])
August 24, 2005
Intelligent Transportation Systems
Institute

Theme:
Human-Centered Technology to Enhance
Safety and Mobility



Scope:
Road- and transit-based transportation
Federally designated University Transportation Center
(Authorized by ISTEA, TEA-21, and reauthorized by
SAFETEA-LU)
Participants in ITS related activities across university:
Over 35 faculty, 14 research staff, ~90 students
Work with Many Organizations
Disciplines from 5 colleges,
including:
• Civil Engineering
• Computer Science
• Electrical Engineering
• Industrial Engineering
• Mechanical Engineering
• Policy and Public Affairs
• Psychology
• Law
…and many counties
Fatalities
100 Million
Vehicle
Miles
Travel
Fatalities
perper
100M
Vehicle
Miles
of of
Travel
Minnesota vs. National
Minnesota
vs. National
Total (1970-2003)
Total (1970-2003)
Minnesota
5
National
4
3
2
1
Year
2003
2000
1997
1994
1991
1988
1985
1982
1979
1976
1973
0
1970
Fatalities per 100M VMT
6
Teen Driving Fatalities: Current Trends



In last decade, have seen an increase in teen fatalities.
Teen drivers have a higher fatality risk than any other driver age
group on the road.
Although teenagers (16-19 years old) make up only 4.7% of all
licensed drivers, they are involved in 13% of all fatal crashes.
Teen (13-19 year old) Fatalities, 1975-2002
10,000
Fatalities
8,000
6,000
4,000
2,000
0
1975
1978
1981
1984
1987
1990
Year
1993
1996
1999
2002
Source: US DOT FARS data.
Teen Crash Risk on a Per Mile Basis
No. of Driver Crash Involvements
per Million Miles Traveled, 1995

Teens are almost twice
as likely to be
involved in a crash
than the next youngest
age group…
20-24 year olds.

16 year olds are nearly
3 times more likely
to be involved in a
crash than a 19 year
old,… and almost 10
times more likely
than drivers aged 3069.
Source: Williams, A. F., 2003. Teenage drivers: patterns of risk. Journal of Safety Research, 34, 5-15.
Driver Fatalities by Crash Type:
For Most Crash Types, Higher for Teens
Driver Fatalities per 100,000 licensed drivers:
Minnesota, 1998 - 2002.
Ran off Road
Right Angle
Head On
Left Turn Oncoming
Rear End
55
50
45
40
35
30
25
20
15
10
5
0
16-19
20-24
30-34
25-29
35-64
65+
Age
Data provided by: Alan Rodgers, Research Analyst for the Minnesota Dept. of Public Safety
Teen Fatality Contributing Factors: Speed Kills
Improper/Unsafe Lane
use
4%
Other
6%
Driver Inexperience
6%
All FATAL Crashes
Illegal/Unsafe Speed
28%
Disregard of traffic
control
7%
Physical Impairment
9%
Driving left of center
11%
Failure to Yield ROW
16%
Driver
Inattention/Distraction
13%
16 TO 19 YEAR OLD DRIVERS: MINNESOTA 1998 - 2002.
Data provided by: Alan Rodgers, Research Analyst for the Minnesota Dept. of Public Safety
Teen Fatality Contributing Factors:
Seatbelt Use
In Minnesota, seatbelt use is lowest among teenagers.
Source: Minnesota Motor Vehicle Crash Facts, 2002
0%
D.C. (1)
Oregon (28)
Hawaii (8)
New Jersey (13)
California (185)
Maryland (32)
Michigan (78)
New Hampshire (9)
Washington (40)
N. Carolina (100)
New York (71)
Iowa (33)
Texas (213)
Illinois (85)
Alaska (6)
Indiana (57)
Total (2526)
Colorado (49)
70%
Wisconsin (62)
Mass (36)
Virginia (57)
Alabama (74)
North Dakota (9)
Nevada (20)
Connecticut (18)
Maine (9)
Ohio (84)
Arizona (45)
Georgia (102)
Vermont (7)
Pennsylvania (118)
Idaho (19)
Florida (151)
Utah (20)
Missouri (97)
Louisiana (60)
Minnesota (53)
Oklahoma (42)
Arkansas (41)
Tennessee (79)
South Carolina (60)
Delaware (10)
Kansas (38)
Mississippi (49)
Montana (21)
Nebraska (21)
South Dakota (12)
Kentucky (49)
West Virginia (22)
Wyoming (6)
New Mexico (20)
Rhode Island (7)
Unrestrained Teen (16-19) Driver Fatalities as
Percent of All Teen Driver Fatalities
(Passenger vehicles only)
Unrestrained Teen (16-19) Driver Fatalities in Passenger Vehicles as Percent
of Total Teen Driver Fatalities in Passenger Vehicles
Source:
FARS 2003
Source: FARS 2003
90%
80%
(x) = Number of Teen Driver Fatalities
60%
50%
40%
30%
20%
10%
80%
Rhode Island (7)
Wyoming (6)
New Mexico (20)
Kentucky (49)
West Virginia (22)
Nebraska (21)
South Dakota (12)
Mississippi (49)
Montana (21)
Delaware (10)
Kansas (38)
Tennessee (79)
S. Carolina (60)
Arkansas (41)
Minnesota (53)
Oklahoma (42)
Missouri (97)
Louisiana (60)
Florida (151)
Utah (20)
Pennsylvania (118)
Idaho (19)
Georgia (102)
Vermont (7)
Ohio (84)
Arizona (45)
Connecticut (18)
Maine (9)
North Dakota (9)
Nevada (20)
Virginia (57)
Alabama (74)
Wisconsin (62)
Mass (36)
Total (2526)
Colorado (49)
Alaska (6)
Indiana (57)
Illinois (85)
Iowa (33)
Texas (213)
N. Carolina (100)
New York (71)
N. Hampshire (9)
Washington (40)
Maryland (32)
Michigan (78)
New Jersey (13)
California (185)
Oregon (28)
Hawaii (8)
D.C. (1)
Unrestrained Teen (16-19) Driver Fatalities as % of All Teen Fatalities,
Unrestrained Non-Teen Driver Fatalities as % of All Non-Teen Driver
Fatalities (Passenger vehicles only)
Unrestrained Teen (16-19) Driver Fatalities as a Percent of Total Teen Fatalities,
Unrestrained Non-Teen Driver Fatalities as Source:
a Percent of Total
Non-Teen
Driver Fatalities (Passenger Vehicles)
FARS
2003
Source: FARS 2003
90%
(x) = Number of Teen Driver Fatalities
Teen Driver
Non Teen Driver
70%
60%
50%
40%
30%
20%
10%
0%
Alcohol Use: For every age group, existing
approaches to mitigation have hit a brick wall
Percent of fatally injured passenger vehicle drivers with BACs >= 0.08 %,
2002
70
60
16-17
18-20
40
21-30
30
>30
20
10
% of all 16-17 year olds who were fatally injured
2002
2000
1998
1996
1994
1992
1990
1988
1986
1984
0
1982
Percent
50
Year
Source data: http://www.hwysafety.org
Behavioral Modification:
Functions
In-vehicle technology does have ability to address these issues by
forcing behavior, providing driver feedback, and reporting driving
behavior of teenagers.
• Forcing Behavior. (“We know better than you.”)
Some unsafe actions (risks) may be habitual. Forcing requires specific
behavior to occur prior to or during vehicle operation.
• Driver Feedback. (Education and adaptation)
Drivers may not be aware of risks. Real-time warnings can alert the
driver in case of poor driving behavior or potential risks.
• Reporting Behavior. (“Big brother is watching”)
Some drivers may purposely take risks because they feel anonymous.
Vehicle parameters can be saved for inspection by parents (or other
authorities).
Three Types of
Forcing Functions

Interlocks (I) force actions in a safe
sequence.

Lock-ins (LI) force an action to be
maintained until it is safe to change.
 Lockout
(LO) blocks unsafe actions.
Forcing Behavior: Interventions
Seatbelt interlock
Requires all occupants to engage seatbelt prior to starting
vehicle. (I)
Alcohol interlock
Prevents teen driver from starting vehicle if alcohol is detected.
(I)
Intelligent Speed Adaptation (ISA)
Prevents driver from exceeding road’s posted limit. Achieved
through combination of Global Positioning System (GPS) and
digital road map. In some systems, speed is limited by link with
elements of vehicle's power train, such as throttle or fuel
system. (LO)
From “Buckle up:
Technologies to Increase Seat Belt Use”
TRB Special Report 278 (2004)



Interlock systems could be engineered to avoid motorists’
objections.
 … e.g. “designed to enable drivers to start cars without
buckling up and to drive in reverse and perhaps at low
speeds to accommodate the majority of drivers who do
not buckle up before starting their vehicles.”
Negative reaction indicated by … NHTSA interviews and focus
groups and hesitancy of industry to reintroduce interlock
systems … suggest that, for the moment, their use be
considered only for certain high-risk groups (e.g., drivers
impaired by alcohol, teenage drivers) who are
overrepresented in crashes.
Current legislation (FMVSS 208; Federal Register 1974, 42,692–
42,693) prohibiting NHTSA from requiring new seat belt use
technologies other than the ineffective 4- to 8-second belt
reminder is outdated and unnecessarily prevents the agency
from requiring effective technologies to increase belt use.
TRB Special Report 278
Recommendations



“NHTSA and the private sector should strongly
encourage research and development of seat
belt interlock systems for specific applications.
“For example, the courts should consider
requiring the use of interlocks for motorists with
driving-under-the-influence-of-alcohol
convictions or with high numbers of points on
their driver’s licenses.
“Interlocks could also be made available for
other high-risk groups, such as teenage drivers.

Insurance companies could lower premium rates for
young drivers who install interlock systems.
On Interlocks and Seatbelt Monitoring
… TRB Special Report 278

“Observational studies indicated that despite their low acceptance,
interlocks were effective in increasing seatbelt use in the 1970’s.
Belt use was 59% in cars with interlocks as compared to 28% in
cars of prior model years. (Robertson, 1975)

“Teenagers have high crash risk but low belt use, which add to
their injury problem. Avenues to address this include:
 Strong belt use laws and their enforcement,
 Building belt use requirements into graduated licensing
systems,
 Keeping young beginners out of high risk driving situations, and
 Finding ways to influence parents and other adults to ensure
that their teenage passengers use seatbelts.” (Williams,
McCartt and Geary, 2003)
Intelligent Speed Adaptation (ISA)
Summary

Three types of ISA systems:




Advisory – in vehicle warning, driver ultimately limits
speed.
Mandatory – active control, vehicle limits speed, overrides
driver.
Voluntary – advisory with option of mandatory.
Three notification levels possible:



Fixed – posted speed limit only.
Variable – site specific limits, ex: construction zones,
school zones, curves.
Dynamic – limits based on hazard potential, e.g. weather,
time of day, traffic congestion, pavement condition.
ISA Summary



Location
 ISA has been evaluated in simulation and field studies in
Australia and several European countries including,
Belgium, France, Germany, England, Netherlands, and
Sweden.
Observations
 In general, these projects have shown consistent
reductions in speed levels, better awareness of speed
limits, and improved compliance with speed limits
(Besseling, 2003; Carsten & Fowkes, 2000; Vagverket,
2003).
Impact
 It has been estimated that speed control systems such as
ISA have the potential for achieving almost 60% fatality
reduction (Carsten & Fowkes, 2000).
“Car computer to stop you speeding”
from The Times, July 1, 2004







Government to establish national speed limits database
… pave way for all cars to be fitted with devices that prevent speeding.
The digital speed map of Britain … essential 1st step towards
introducing ISA, … automatically applies brakes or blocks accel.
On-board computer linked to satellite positioning system will use digital
map to identify local speed limit. If drivers attempt to exceed limit, they
hear series of bleeps and accelerator pedal starts vibrating.
Ministers have not ruled out eventually making some version of system
compulsory ….
…but no central speed limits database for whole country, and many
local authorities have poor records of limits on their roads.
The DfT believes the absence of a national database is hampering
development of ISA.
A DfT spokesman said: “If the whole country was mapped, it might
make it more logical and practical for manufacturers to consider
offering ISA. There could well be road safety benefits from ISA.”
ISA: Compensation for “Lost Time”
No ISA
Based
on speed limits
8
Based
on local cond’ns
Gap
7
Min ttc
Seconds
6
5
4
3
2
1
0
Baseline
Driver Select
Mandatory
Variable
When drivers have speed restricted by a mandatory or variable system, there may be a
tendency for them to compensate by accepting shorter gaps in crossing traffic and closer
following distances in traffic compared to baseline driving (or only an advisory system).
This is believed to result from a perceived need to make up for limited mobility and time.
ISA: Complacency
We relax our responsibility and let the system take over
Adaptive Speed Control
120
110
Fog
warning
100
90
80
70
Distance along section (m)
1100
1000
900
800
700
600
500
400
300
200
100
60
0
Speed (km/h)
In expt, subjects drove a
simulator in traffic
conditions with heavy fog.
Note that drivers drove at
Speed limit
a speed lower than the
speed limit with no speed
Mand, fixed
control, and in response to
the advisory and dynamic
system. However, drivers
NSL (off)
with the mandatory
Advisory
system seemed to be
Mand, dynamiccomplacent and drive to
the system limit rather
than use their own
judgment to slow down.
As a result, drivers tended
to drive toward the speed
limit even though
conditions suggest a lower
speed to be safe.
(courtesy of O. Carsten, ITS, Leeds)
Alcohol Interlock

Commercially available
option





Expensive (?): $795 or
$60/mo.
Records data log of tests,
and rolling retests.
Interlock tolerance level
can be changed.
Installed by certified
dealer.
Integrate with system.
ADS Determinator
Interlock
Manufacturers
Guardian Interlock (www.guardianinterlock.com)
Autosense International (San Jose, CA)
Consumer Safety Technology, Inc. (www.intoxalock.com)
Draeger Interlock, Inc. (www.ignitioninterlock.com)
Lifesaver Interlock, Inc. (www.lifesafer.com)
Smart Start, Inc. (www.smartstartinc.com)
Alcohol Countermeasures Systems (www.acs-corp.com)
Alcohol Detection Systems (www.stopdwi.com)
Enabling Technology:
Smart Key
Benefits:

Enables the system to
recognize who is driving so
parents can opt out.

Enables individual settings
for parents of multiple teen
drivers.

Enables logging of
Graduated Driver’s License
(GDL) provision on number
of driving hours (day/night)
logged with a parent.
Microlatch FOB-13
$295
Features:



Biometrics fingerprint key
fob.
Wireless (bluetooth)
communication.
Enables identification and
vehicle entry.
Modifying Behavior: Feedback
Need context (static and dynamic)
 Auditory or other sensory signals
triggered by unsafe vehicle operation

Excessive speed for local conditions, e.g.
speeds incompatible with road curvature, can
lead to lane departure.
 “Hassles” driver until behavior is corrected.


Prediction of road curvature can inform
the driver of necessary upcoming
maneuvers (especially useful in rural
areas at night).
Reporting Behavior: Consequences
Incentives, Reward and Punishment

Record vehicle parameters such as speed,
acceleration, braking, throttle use, distance,
time of day.

Parents can be notified in real-time of unsafe
driving behavior. Parents can also inspect
“report card” of data to review teen driving
behavior offline.

Attempt to address difficulty in enforcing
compliance. Review possible by insurance
(insurance premium, rebates), police (fines),
DPS (license progression, awards).
Design Opportunities
Indirectly
Speed
Inexperience
Inattention/ Alcohol
Distraction
Forcing
X
Feedback
X
X
X
Reporting
X
X
X
Seatbelts
X
X
X
X
When needed?
Crash rate by cumulative miles driven
after licensure and by gender


First 250 miles crash
involvement rate: 3.2
(per 10K miles); next
250 miles rate is 1.3
(per 10K miles) (1)
For novice drivers,
crash rates decrease
dramatically from the
1st to the 7th month
(41%), then gradually
decrease through the
24th month after
licensing (60% overall
reduction) (2)
(1)
(2)
Mayhew, D.R., Simpson, H.M. and Pak, A. (2003).
“Changes in collision rates among novice drivers during the first months of driving.”
Accident Analysis and Prevention, 35, pp. 683-691.
McCartt A.T.; Shabanova V.I.; Leaf W.A. (2003).
“Driving experience, crashes and traffic citations of teenage beginning drivers,”
Accident Analysis and Prevention, 35, (3), pp. 311-320
Davis Instruments: CarChip

Summary
 Records driving data saved for later viewing on home PC.
• Time and date for each trip, distance, speed, hard
accelerations and decelerations.
 Data logger will start collecting data as soon as car is started.
 Connected via OBD-II port (available on model years 1996+ )
 Cost: $179.
Road Safety: RS-1000

Summary
 Records driving characteristics for later viewing on PC by
guardian.
 Audible alert parameters defined by user.
• Speed, driver seatbelt, acceleration, braking, erratic driving,
throttle use.
 Connects via OBD-II port (not seatbelt monitor)
 GPS plug-in option available soon (?)
 Relatively low cost: $280
Geofencing: SignalTrac
http://www.signaltrac.com/

Summary
 Real-Time e-mail notification via cellular connection each time
driver enters a zone.
 Zone(s) are user-defined.
 Also includes real-time position, speed, seatbelt*, and
passenger* notification.
(*Optional)
Cost: $499 + $399/yr
Teen Arrive Alive

Summary
 Subscription plan for phone tracking.
 Uses technology from GPS enabled cell
phone.
 Works with selected Motorola phones
and Nextel calling plans.
 Subscription cost: ~$20/month (in
addition to standard Nextel service plan
fees of ~$40/month).
 Phone location, speed, direction of travel,
and time of day are reported every 2
minutes.
 Reports are accessible by parents via
website or by placing a call to secure line.
Video Monitoring: DriveCam
http://www.drivecam.com/drivecam-videos.asp
Features:

Two lenses: Forward and Interior.
20 second buffer records 10s prior to,
and 10s after event.


Records both Video & Audio.
Limitations:
Cost: $1000

User defined threshold - false positives.

Difficult to record speed or impairment.
Feedback to driver behavior provided after
event (not real-time)


Review of footage is time consuming.
Existing teen driving aids
Product
Manufacturer
website
base price ($)
activation fee ($)
service fee ($)
Data Collected
video
location (GPS)
speed
distance
acceleration
deceleration/braking
lateral acceleration
throttle position
time of day
driver seatbelt use
alcohol
unsafe backing
Notification
real-time GPS
in-vehicle display
online reports
e-mail alerts
cellphone alerts
Additional Features
RS-1000
RoadSafety
CarChip E/X DriveRight 600 SignalTrac NetworkCar SmartDriver
Davis Instruments
SignalTrac NetworkCar SmartDriver
roadsafety.com
davisnet.com
TeenArriveAlive
TeenArriveAlive
drivecam.com
teenarrivealive.com
280
N
N
179
N
N
395
N
N
499
145
399/yr
995
90
108/yr
495
69
100/yr
1,200
N
N
N
50
240/yr
N
R**
FE
Y
FE
FE
FE
R
R
FE*
N
FE*
N
N
R
R
R
R
N
R
R
N
N
N
N
R
FE
R
FE
FE
N
N
R
R*
N
N
N
R
R
R
N
N
N
N
R
R*
N
N
N
R
N
N
N
N
N
N
R
N
N
N
N
N
R
R
N
N
N
R
R
N
N
N
R
N
N
N
R
R
R
N
N
R
N
N
N
R
R
R
N
N
N
N
R
N
N
N
R**
N
N
N
N
1,3
N
N
N
N
N
N
FE
N
N
N
2,3
R
N
R
R
N
1
R
N
R
N
N
1
N
N
N
N
N
N
N
N
N
N
R
N
N
N
R
1
FUNCTIONS
FO - FORCING
FE - FEEDBACK
R - REPORTING
N - NOT AVAILABLE
signaltrac.com netw orkcar.com smart-driver.com
DriveCam
DriveCam
*feature is optional
**feature is not yet available
(1) broadcasts via cellular connection
(2) LCD display on dashboard
(3) auditory feedback
Existing system deficiencies
Also see www.gps-practice-and-fun.com/teen-car-tracking.html

Current systems are too passive. None of the systems:
 Modify speed threshold based on individual road’s local
speed limit, or upcoming road curvature (as per ISA),
time of day or weather (RWIS).
 Force behavior such as using seatbelt or maintaining
sobriety.
 Recognize current driver.

Digital maps can be updated with speed limits; real-time
wireless access for pavement condition, weather, congestion
already available in Minnesota.


Minnesota: One point of contact for all statutory speed limits.
Need teen driver-parent centric system, designed to modify
dangerous teen driving behavior and empower parents.
Driver Reporting Systems:
The issue is not only “technology”






What are the tests? The performance criteria?
 Speed violation? Stability of accel/decel, headway?
Lane wandering? Distraction measure?
What thresholds does one set for pass/fail on each?
How does one come up with an overall “grade”?
Is this a continuous driving exam? What are the
thresholds for moving from one level to the next?
Does one exam (ie report card) fit every state? …every
teen?
Feedback mechanism? Incentive?
Mechanisms of Unsafe Driving:
“Reporting” cannot deal with all of these

Perception:


Recognition:


Youth and personality (sensation seeking) may attract teen driver to thrill of
risk taking and unsafe speed
Motivation:


Insufficient experience to acquire adequate speed control skills.
Personality:


Insufficient experience to recognize unsafe limits.
Skill:


Insufficient experience to accurately perceive speeds.
Absence of external factors to motivate (“enforce”) safe speeds. Anonymity.
Peer pressure motivates risky behavior.
Naivety:

Absence of sufficient exposure to negative consequences of speed choice
to “learn” risks of unsafe speeding; optimism bias
Mechanisms of Unsafe Driving:
Beyond “Reporting”
Forcing
Feedback
Reporting
Perception
Experience
Awareness
Perception
Experience
Skill Awareness
Personality
Motivation
Awareness
Mechanisms:






Perception
Recognition
Skill
Personality
Motivation
Naivety

Ignition interlocks
to deal with impairment
(which affect Perception,
Recognition, & Skill)
Technical Caveats

Technology can be provided at different levels of
intervention.



Inform
Advise, warn
Control

Technology may redistribute risk rather than reduce risk
overall.

By changing the nature of the driving/operational task, we
need to verify that the technology not introduce new risk
factors.

The question always is: What are the unintended
consequences of using technology?
Driving Simulator and Research Tools
Special Effects:
Night-time
capabilities