DADSS Transportation Research Board 90th Annual Meeting 01-25-11
-
Upload
kaye-beach -
Category
Documents
-
view
56 -
download
2
description
Transcript of DADSS Transportation Research Board 90th Annual Meeting 01-25-11
WWW.DADSS.ORG
Transportation Research Board
90th Annual Meeting
January 25, 2011
Washington, DC
Bud Zaouk, D.Sc.
QinetiQ North America
Driver Alcohol Detection System for Safety
DADSS
WWW.DADSS.ORG
2
Driver Alcohol Detection System for Safety
WHY DO WE NEED DADSS AND HOW WE GOT HERE
WWW.DADSS.ORG
Why do we need DADSS
.00-.049 BAC ���� 77%
1.5 Million DWI Arrests Annually
1 Million DWI Convictions Annually
150,000+ Ignition Interlocks Installed
.05-.079����12% .08+����11%
906 Million driving trips within two hours of consuming alcohol
94 Million
233 Billion Trips Annually by Car and Light TruckFHWA/NHTS
2009
NHTSA - 2001
WWW.DADSS.ORG
Why do we need DADSS
0
100
200
300
400
500
600
700
800
0.0
1
0.0
4
0.0
7
0.1
0.1
3
0.1
6
0.1
9
0.2
2
0.2
5
0.2
8
0.3
1
0.3
4
0.3
7
0.4
0.4
3
BAC
Nu
mb
er
of
Dri
vers
Distribution of BAC Levels for Drivers involved in Fatal Crashes With a BAC of .01 or Higher, 2007
NHTSA, 2008
WWW.DADSS.ORG
5
Driver Alcohol Detection System for Safety
COOPERATIVE AGREEMENT
WWW.DADSS.ORG
Driver Alcohol Detection System for Safety
♦ The Automotive Coalition for Traffic Safety and NHTSA entered into a cooperative agreement in February 2008 to “explore the feasibility, the potential benefits of, and the public policy challenges associated with a more widespread use of unobtrusive technology to prevent drunk driving”
♦ Goal is to develop non-invasive, seamless technologies to measure driver BAC and reduce the incidence of drunk driving
♦ Systems need to measure alcohol accurately, precisely, reliably, and in a very short time so the sober driver is not inconvenienced
♦ Five-year program to develop and test prototypes that may be considered for vehicle integration thereafter
♦ Devices intended to prevent alcohol impaired drivers (BAC ≥ 0.08) from driving their vehicles
6
WWW.DADSS.ORG
Participating Manufacturers
7
WWW.DADSS.ORG
DADSS Blue Ribbon Panel
♦ BRP appointed by ACTS and works in an advisory capacity
♦ Comprised of experts from various disciplines, including
� Auto manufacturers
� Suppliers
� Alcohol toxicology
� Impairment
� Ignition interlocks
� Human factors
♦ BRP assigned working groups to assist in effort
� Technical Working Group
� Public Acceptance and Public Policy Working Group
� Research scientists
� MADD
� IIHS
� NHTSA
� Foreign governments
WWW.DADSS.ORG
9
2008 2009 2010 2013
Phase I Funding
Phase II Funding
Perform Interior Mockup Testing
Develop DADSS Research Vehicle
Implement DADSS Subsystem(s) in Vehicle
Interior Mockup
Perform DADSS Research Vehicle
Testing
Phase II Subsystem Development
Human Subjects Tests
Human Subjects Tests
Interior Mockup Tests
Vehicle Tests
Perform Technology Verification
Perform Prototype(s) Lab Testing
Develop DADSS
Subsystem Prototype(s)
BenchTests
Phase I Prototype
Development
Human Subjects Tests
DADSS Program Process
Q3Q2
Patents and Literature Review
Performance Specifications
Request for Proposals
Assess Current State of Technology
Request for Information
WWW.DADSS.ORG
10
Driver Alcohol Detection System for Safety
PROGRAM TIMELINE
WWW.DADSS.ORG
Program Timeline
2008 – H1 2008 – H2
CRA
Signed
RFI
ReleasedPMP
Approved
PS Draft7
Released
RFP
Released
RFI
Responses
Task 3
Report
RFI Evaluation
and on-site
visits
2009 – H1 2009 – H2
Offerors
Presentations
RFP Responses
RFP Evaluation
Task 4
Report
Phase I
Awards
SCD Development
PS Draft8 Released
11
WWW.DADSS.ORG
Program Timeline
12
2010– H1 2010 – H2
Phase I
Breath SCD
Phase I
Touch SCD
Research
Vehicle
Integration
Requirements
Phase II
RFP
Prototypes Evaluation
Human
Subjects
Test Protocol
WWW.DADSS.ORG
13
Driver Alcohol Detection System for Safety
UNDERSTANDING ALCOHOL ABSORPTION AND ELIMINATION
WWW.DADSS.ORG
14
Stomach
Intestines
Liver Heart
Lungs
Capillary Bed
Tissue and OrgansSweat/Vapor
Waste
Arterial BAC
Inferior Vena Cava
Hepatic Artery
Venous
BAC
Art
eri
al
Blo
od
Ven
ou
s
Blo
od
Breath Alcohol Concentration
Transdermal Alcohol Concentration
Tissue Alcohol Concentration
Urine
Alcohol Conc.
AbsorptionAlcohol and Elimination
Portal Vein
WWW.DADSS.ORG
15
Driver Alcohol Detection System for Safety
OVERVIEW OF PHASE I TECHNOLOGIES BEING FUNDED FOR DEVELOPMENT
WWW.DADSS.ORG
% T
ran
smit
tan
ce
3 3.5 4 5 6 7 8 1510 209.4
% T
ran
smit
tan
ce
3 3.5 4 5 6 7 8 1510 209.4
DADSS Technologies
Distant Spectrometry
♦ Breath-based system
♦ Close correlation between BrAC and arterial BAC
♦ Extensive real world experience
Tissue Spectrometry
♦ Touch-based system
♦ Extensive human subjects testing done
♦ BAC measurements closer to capillary (arterial) blood than to breath
Visible Near Infrared
TruTouch
AFRICAN AMERICAN
CAUCASIAN
Visible Near Infrared
TruTouch
AFRICAN AMERICAN
CAUCASIAN
16
WWW.DADSS.ORG
Autoliv Phase I Prototype
Breathing Cup
Pump Filter
L=6.25 in.W=3.25 in.H=3.75 in.
Emitter
SensorsSample
InletOptical Module
Reflecting Mirrors
♦ Autoliv prototype uses� Alcohol (ethanol) and carbon dioxide
measured by Infrared spectroscopy sensor• Carbon dioxide allows measurement of
breath dilution
• Unobtrusive “sniffer” to detect alcohol in the vehicle
� Multiple sensors in-vehicle (steering wheel, A-Pillar, etc..)
♦ To obtain necessary accuracy and resolution � Driver may have to deliver a forced
expiration towards the sensor� Would be needed in less than 1
percent of cases to ascertain whether above or below legal limit
WWW.DADSS.ORG
Autoliv In-Vehicle Sensing
♦ In-vehicle expired breath aerodynamics � Estimate carbon dioxide levels at various
positions � Ventilation on and off� Windows opened or closed
Door closing
<5 sec after door closure for sensors on
steering wheel
In-Vehicle Signal Pattern
Door closing
<5 sec after door closure for sensors on
steering wheel
In-Vehicle Signal Pattern
Door1
closing
Door1
opening
Vehicle
unlocking
OFF STDBY ZERO ACTIVE
VEHICLE DISABLED
VEHICLE DISABLED
VEHICLE ENABLED
STDBY
CO2
EtOH
5 sec
”YELLOW ZONE”
TIME
WWW.DADSS.ORG
Alcohol Countermeasure Systems
♦ ACS prototype uses mid infrared detection methodology � Broadly tunable Daylight Solution’s
External Cavity Quantum Cascade Laser (ECqcL™)
♦ ECqcL™� Ability to capture entire ethanol
spectrum� Allows detection of interferents
♦ Carbon dioxide allows measurement of breath dilution
Measurement Conditions
100 sweeps10 sec. acquisition time
WWW.DADSS.ORG
TruTouch Technologies
♦ TruTouch uses near infrared to measure tissue alcohol concentration (TAC) in the Dermis
♦ Outside of index finger used� Palmar side of fingers will be
used in Phase II
20
stratum corneumstratum corneum
epidermisepidermis
dermisdermis
subcutaneoussubcutaneous
Light Source
Optical Touchpad
Interferometer Engine
Processing Electronics
WWW.DADSS.ORG
21
Driver Alcohol Detection System for Safety
PERFORMANCE SPECIFICATIONS AND
STANDARD CALIBRATION DEVICES
WWW.DADSS.ORG
DADSS Phase I Requirements
♦ Phase I POP Prototypes evaluated against the following performance specifications:� Measure from 0.01% to 0.12% BAC
� Measurement time = 325 milliseconds
� Accuracy and Precision• 0.07%-0.09% BAC � ±0.0003% BAC
• Requires Standard Calibration Devices (SCD)– Breath-based systems
– Tissue-based systems
DADSS Performance Specifications DRAFT08
available at http://www.dadss.org
High accuracy but low precision
High precision but
low accuracy
% BACDADSS
AccuracyEvidentialAccuracy
DADSS Precision
EvidentialPrecision
0.010 - 0.050 0.0010 0.0050 0.0010 0.0042
0.050 - 0.070 0.0007 0.0050 0.0007 0.0042
0.070 - 0.090 0.0003 0.0050 0.0003 0.0042
Greater than 0.090 0.0010 0.0050 0.0010 0.0042
More accurate calibration source required for DADSS program
WWW.DADSS.ORG
Standard Calibration Devices (SCD)
Objective
♦ Assess and document the accuracy and precision of the Phase I Proof-of-Principal (PoP) prototypes
Approach
♦ Provide sample sources of “breath”or “tissue” (SCD) to PoP sensor
� Known and consistent alcohol content
Challenge
♦ Develop process to assure the SCD performance
� Delivers targeted samples to PoPsensor
� Has to exceed accuracy and precision requirements
23
Wet Gas Breath Alcohol
Simulator Dry Gas ±0.5 ppm(±0.0002 %BAC)
Tissue SCD Delivery System
Tissue SCD
WWW.DADSS.ORG
Wet Gas Breath Simulator
24
3000 p
pm
EtO
H
An
d N
2 M
ix
Humidifier100% RH
Pressure Vessel
Hygrometer to measure
humidity
Vacuum Pump
34°°°°
C
Environmental Chamber Temp.
controlled at 45°°°°C
MFC 1MFC 2
6%
CO
2/ 16%
O2
/
78%
N2
Gas Blender
♦ Models human breath volume
and pressure profile
♦ Uses pressurized vessel with
humidified gases
♦ Adjustable level of humidity
(80-100% RH)
♦ Adjustable alcohol
concentration (gas blender)
♦ Process heated to prevent
condensation
♦ Ethanol introduced post
humidification
♦ Ability to meet DADSS
Specification
5L
1.5 sec
WWW.DADSS.ORG
SCD – Tissue-Based Systems
♦ Electromechanical fluidic system to introduce samples to sensor
♦ Mimic average optical scattering properties of human skin� Incorporates NIR “reflective”
polymer beads (microspheres) � Simulates collagen in tissue (in-
vitro scattering samples)
♦ Varied over range of concentrations
♦ Potential to meet DADSS Specs
25
Compound Name
Base Reagent Reagent Function
AlbuminBovine Serum
AlbuminSimulator of blood density
Creatinine Creatinine Component within blood
Ethanol200 proof HPLC
Grade-
SalinePhosphate Buffered
SalineAdds salt and adjusts pH
Triton Triton X-100Prevent microspheres
from clumping
Urea Urea Components within blood
Water ACS Reagent Water Mixing agent
MicrospheresSimulate collagen NIR
reflectance and scattering
4500 5000 5500 6000 6500 7000 7500
-8
-7.5
-7
-6.5
-6
-5.5
-5
Wavenumber (cm-1)
2% Bead Solution
Human Finger
(μm) 2.22 2.00 1.81 1.66 1.53 1.4 1.33 1.258000
Electromagnetic Spectrum
Bead Solution Compared to Finger Spectra
Pse
ud
o-A
bso
rba
nce
(cm-1)
4500 5000 5500 6000 6500 7000 7500
-8
-7.5
-7
-6.5
-6
-5.5
-5
Wavenumber (cm-1)
2% Bead Solution
Human Finger
(μm) 2.22 2.00 1.81 1.66 1.53 1.4 1.33 1.258000
Electromagnetic Spectrum
Bead Solution Compared to Finger Spectra
Pse
ud
o-A
bso
rba
nce
(cm-1)
WWW.DADSS.ORG
26
DADSS Program Review
POP PROTOTYPES BENCH TESTS EVALUATION
WWW.DADSS.ORG
Bench Test Evaluations
♦ Bench Test Objectives
� Independent evaluations of delivered PoP Prototypes
� Measurement of system performance based upon
• Time to present alcohol reading
• Accuracy of reading
• Precision of reading
� Bench Test Protocols developed for
• Breath based
• Tissue based
27
WWW.DADSS.ORG
Touch-Based Bench Test Results
♦ At 0.08 %BAC prototype exceeded DADSS accuracy at the longer measurement times
♦ At shorter measurement times estimated accuracy somewhat higher than specifications
♦ Precision estimates fell short of the DADSS specifications at all sampling periods
♦ Represents significant progress compared with specifications for existing evidential breath-test devices
WWW.DADSS.ORG
TruTouch Bench Test Results at 0.080 % BAC
29
0.0001
0.0011
0.0001
0.0016
0.0005
0.0021
0.0013
0.0043
0.0000
0.0010
0.0020
0.0030
0.0040
0.0050
0.0060
Eth
an
ol co
ncen
trati
on
(%
BA
C)
60 sec. 30 sec. 15 sec. 5 sec.
SD
60 sec. 30 sec. 15 sec. 5 sec.
SE
DADSS SE & SD =
0.0003 % BAC
0.0001
0.0011
0.0001
0.0016
0.0005
0.0021
0.0013
0.0043
0.0000
0.0010
0.0020
0.0030
0.0040
0.0050
0.0060
Eth
an
ol co
ncen
trati
on
(%
BA
C)
60 sec. 30 sec. 15 sec. 5 sec.
SD
60 sec. 30 sec. 15 sec. 5 sec.
SE
DADSS SE & SD =
0.0003 % BAC
Evidential SE =
0.0050 % BAC
Evidential SD =
0.0042 % BAC
Accuracy Precision
WWW.DADSS.ORG
Autoliv Bench Test Results
♦ Measurement time of 5 seconds used in all tests
� Dry gas released for 10 Seconds
� Data collected for 5 Seconds
♦ Source directly connected to sensor with ¼”length tube
♦ Accuracy estimates were close to or below the DADSS specifications
♦ Precision estimates exceeded the specifications, but were still lower (better) than evidential standards
WWW.DADSS.ORG
0.00020.0004
0.0000
0.0017
0.0022
0.0027
0.0000
0.0010
0.0020
0.0030
0.0040
0.0050
0.0060
0.020 % BrAC 0.080 % BrAC 0.120 % BrAC
Eth
an
ol
Co
nce
ntr
ati
on
(%
BrA
C)
Accuracy
Precision
DADSS SE & SD =
0.0003 % BAC
0.00020.0004
0.0000
0.0017
0.0022
0.0027
0.0000
0.0010
0.0020
0.0030
0.0040
0.0050
0.0060
0.020 % BrAC 0.080 % BrAC 0.120 % BrAC
Eth
an
ol
Co
nce
ntr
ati
on
(%
BrA
C)
Accuracy
Precision
DADSS SE & SD =
0.0003 % BAC
Autoliv Direct Dry Gas Bench Test Results
31
Evidential SE =
0.0050 % BAC
Evidential SD =
0.0042 % BAC
WWW.DADSS.ORG
ACS Bench Test Results
♦ Results of ACS prototype yielded ambiguous results
� Prototype sensor, hardware, and software extremely
cumbersome to work with
� More difficult and time consuming to operate so very few
tests completed with high variability
♦ Prototype failed to meet the DADSS specifications for
accuracy and precision at all BrAC concentrations
� Was not able to meet the current evidential standards
32
WWW.DADSS.ORG
0.0045
0.07190.0676
0.0085
0.0193
0.0573
0.000
0.010
0.020
0.030
0.040
0.050
0.060
0.070
0.080
0.090
0.100
0.020 % BrAC 0.080 % BrAC 0.12 % BrAC
Eth
an
ol
con
cen
tra
tio
n (
% B
rAC
)
Accuracy
Precision
0.0045
0.07190.0676
0.0085
0.0193
0.0573
0.000
0.010
0.020
0.030
0.040
0.050
0.060
0.070
0.080
0.090
0.100
0.020 % BrAC 0.080 % BrAC 0.12 % BrAC
Eth
an
ol
con
cen
tra
tio
n (
% B
rAC
)
Accuracy
Precision
ACS Bench Test Results
33
Evidential SE =
0.0050 % BAC
Evidential SD =
0.0042 % BAC
WWW.DADSS.ORG
34
DADSS Program Review
HUMAN SUBJECT TESTS
WWW.DADSS.ORG
Human Subject Testing
♦ Subjects dosed to reach a BAC of 0.12 g/dL ♦ Test procedures
� Blood is drawn at a rate of 1 ml/min� Samples taken every 2.5 minutes
♦ Every 5 minutes subjects provide� Evidential breath sample� Short puff of breath into Autoliv Prototype� Presses finger on touch pad of TruTouch prototype
35
0.000
0.020
0.040
0.060
0.080
0.100
0.120
0.140
0.160
0.180
0.200
0:00 0:30 1:00 1:30 2:00 2:30 3:00 3:30 4:00 4:30 5:00 5:30 6:00 6:30 7:00
Alc
oh
ol co
ncen
tra
tion
(g
m%
)
Elapsed time (h:mm)
DADSS Phase I Human Subject Test 5 (ASQ009) August 26, 2010
TruTouch (BAC)
Evidenzer (BrAC)
Autoliv R2 (BrAC)
Whole Blood (BAC)
Dosing
Blood Clot
0.000
0.020
0.040
0.060
0.080
0.100
0.120
0.140
0.160
0.180
0.200
0:00 0:30 1:00 1:30 2:00 2:30 3:00 3:30 4:00 4:30 5:00 5:30 6:00 6:30 7:00
Alc
oh
ol co
ncen
tra
tion
(g
m%
)
Elapsed time (h:mm)
DADSS Phase I Human Subject Test 5 (ASQ009) August 26, 2010
TruTouch (BAC)
Evidenzer (BrAC)
Autoliv R2 (BrAC)
Whole Blood (BAC)
Dosing
Blood Clot
0.000
0.020
0.040
0.060
0.080
0.100
0.120
0.140
0.160
0.180
0.200
0:00 0:30 1:00 1:30 2:00 2:30 3:00 3:30 4:00 4:30 5:00 5:30 6:00 6:30 7:00
Alc
oh
ol co
ncen
tra
tion
(g
m%
)
Elapsed time (h:mm)
DADSS Phase I Human Subject Test 11 (ASQ-016) November 04, 2010
TruTouch (BAC)
Evidenzer (BrAC)
Autoliv R2 (BrAC)
Whole Blood (BAC)
Dosing
0.000
0.020
0.040
0.060
0.080
0.100
0.120
0.140
0.160
0.180
0.200
0:00 0:30 1:00 1:30 2:00 2:30 3:00 3:30 4:00 4:30 5:00 5:30 6:00 6:30 7:00
Alc
oh
ol co
ncen
tra
tion
(g
m%
)
Elapsed time (h:mm)
DADSS Phase I Human Subject Test 11 (ASQ-016) November 04, 2010
TruTouch (BAC)
Evidenzer (BrAC)
Autoliv R2 (BrAC)
Whole Blood (BAC)
Dosing
WWW.DADSS.ORG
36
Driver Alcohol Detection System for Safety
PUBLIC ACCEPTANCE AND POLICY ISSUES
WWW.DADSS.ORG
Acceptance among the public and key leaders is critical
♦ Technology will be effective only if the driving public welcomes and accepts it:
� 58 percent of the U.S. public say they support smart technology to prevent driver impairment including alcohol-impaired driving (MADD U.S survey, 2006)
� 64 percent of the U.S. public say they support advanced technology in all vehicles, if it is reliable, to prevent anyone with an illegal BAC from driving their car (IIHS, 2009)
37
WWW.DADSS.ORG
Public Acceptance Research Needs
♦ Assess what technology solutions might prove the most
acceptable and how they might best be implemented
♦ Assess current levels of consumer understanding of
DADSS, and levels of acceptance
♦ Monitor public acceptance levels over time and
willingness to adopt the technology on their vehicles
♦ Educate the public about potential technological
solutions as it relates to the alcohol-impaired driving
problem
♦ Have discussions about policy issues with relevant
stakeholder groups
38
WWW.DADSS.ORG
Planned Public Acceptance Research
♦ Focus Groups – To be conducted February-April 2011
� Gauge public perception of advanced in-vehicle alcohol
detection technologies, and identify potential barriers to their introduction
� Provide input to technology design of the two approaches
currently being taken
� Seek guidance on technology introduction strategy into the
passenger vehicle fleet
� 4 locations within the U.S.
• Boston, MA - Portland, OR - Albuquerque, NM - Milwaukee, WI
� 3 focus groups at each location
• Non-drinkers, social drinkers, and heavy drinkers
WWW.DADSS.ORG
Planned Public Acceptance Research
♦ Periodic national surveys � Gauge public understanding and attitudes towards DADSS systems
in all vehicles,
� Assess the public’s reactions to potential technology features to aid the technology development process, and
� Assess their state of knowledge regarding drinking and the relationship to impairment, drinking and driving and legal thresholds, etc
♦ Stakeholder group discussions
� Increase stakeholder understanding of and potential support for the integration of DADSS devices into the vehicle fleet
� Seek input among groups not currently included in BRP that could be used to inform the development of technology
♦ Review and revise DADSS Performance Specifications, as may be indicated by the findings of the public acceptance research
WWW.DADSS.ORG
Communicating with the Public
♦ A website has been launched to provide public information:
Content
� A “go to” site to provide project details,
� Answer key questions about technology development and drinking and driving
www.dadss.org
41
WWW.DADSS.ORG
Summary
♦ DADSS technologies have been identified, contracts awarded, and prototypes have completed bench and human subjects testing
♦ Preliminary phase I results indicate there are technologies showing sufficient promise to meet DADSS Performance Specifications withrespect to measurement time, accuracy, and precision
♦ Technology developers have identified the work needed to meet the DADSS requirements (gap analysis)
♦ A public acceptance research plan has been developed and is underway – the public must be knowledgeable about the system and see its benefit in their vehicles
♦ Dialogue continues with policy makers and other key stakeholdersto ensure their support
42
WWW.DADSS.ORG
Next Steps
43
2011– H1 2011 – H2
Awardees
Selection
RFP
Evaluation
Phase II Kick-
Off-Meetings
Begin Research
Vehicle
Development
Phase II
RFP
On-Site Progress
Report
On-Site Progress
Report
WWW.DADSS.ORG
44
Driver Alcohol Detection System for Safety
QUESTIONS?
http://www.dadss.org
Contact Information
Bud Zaouk