VOTERS - Northeastern University · • Automatic quantitative pavement surface analysis, indicates...

29
VOTERS Versatile Onboard Traffic Embedded Roaming Sensors Overview End of Year One NIST Site Visit Ming L. Wang, Director This work was performed under the support of the U.S. Department of Commerce, National Institute of Standards and Technology Technology Innovation Program Cooperative Agreement Number 70NANB9H9012 Northeastern University Feb. 25 th , 2010 | 1 www.neu.edu/voters PROPRIETARY Technology , Technology Innovation Program, Cooperative Agreement Number 70NANB9H9012

Transcript of VOTERS - Northeastern University · • Automatic quantitative pavement surface analysis, indicates...

Page 1: VOTERS - Northeastern University · • Automatic quantitative pavement surface analysis, indicates pavement health (surface profile and crack density/porosity), ice, water, oil cover.

VOTERSVersatile Onboard Traffic Embedded 

Roaming Sensorsg

OverviewEnd of Year One NIST Site Visit

Ming L. Wang, Director

This work was performed under the support of the U.S. Department of Commerce, National Institute of Standards and Technology Technology Innovation Program Cooperative Agreement Number 70NANB9H9012

Northeastern University

Feb. 25th, 2010  | 1 www.neu.edu/votersP R O P R I E T A R Y 

Technology, Technology Innovation Program, Cooperative Agreement Number 70NANB9H9012

Page 2: VOTERS - Northeastern University · • Automatic quantitative pavement surface analysis, indicates pavement health (surface profile and crack density/porosity), ice, water, oil cover.

Overview OutlineOverview Outline

VOTERS Vi i• VOTERS Vision

• Management

• Technical Highlights

• OutreachOutreach

• Summary

Feb. 25th, 2010  | 2 www.neu.edu/votersP R O P R I E T A R Y 

Page 3: VOTERS - Northeastern University · • Automatic quantitative pavement surface analysis, indicates pavement health (surface profile and crack density/porosity), ice, water, oil cover.

Vi iVision

Reliable, safe, and cost effective highwayeffective highway maintenance system

A systems approach

Feb. 25th, 2010  | 3 www.neu.edu/votersP R O P R I E T A R Y 

Page 4: VOTERS - Northeastern University · • Automatic quantitative pavement surface analysis, indicates pavement health (surface profile and crack density/porosity), ice, water, oil cover.

Reasons for Infrastructure Vulnerability

• Poor or incomplete knowledgeknowledge

• Lack of data prevents   critical decision making

• Current condition inCurrent condition in disrepair

• Lack of fund to repair

Feb. 25th, 2010  | 4 www.neu.edu/votersP R O P R I E T A R Y 

Page 5: VOTERS - Northeastern University · • Automatic quantitative pavement surface analysis, indicates pavement health (surface profile and crack density/porosity), ice, water, oil cover.

Benefits of VOTERS SystemBenefits of VOTERS System

• Added value of VOTERS to existing and improved g pinfrastructure management systems

• Goal to make maintenance decision easier through more f l i f tiuseful information

– Denser spatial and temporal coverage

– Measurements providing useful information currently not usedp g y

• If some information (e.g. potholes, poor roadways) made public, it could make entities in charge of planning and management of maintenance more acco ntablemanagement of maintenance more accountable

• Raise awareness of how cost effects road and bridge conditions

Feb. 25th, 2010  | 5 www.neu.edu/votersP R O P R I E T A R Y 

Page 6: VOTERS - Northeastern University · • Automatic quantitative pavement surface analysis, indicates pavement health (surface profile and crack density/porosity), ice, water, oil cover.

• Innovative Solutions for Infrastructure Management

• Use Vehicles of Opportunity as economic and safe fleet of sensor systems

• Collect reliable geo‐referenced multi‐sensor data sets with dense spatial and temporal coverage

• Provide added value information for pavement and bridge deck management

Feb. 25th, 2010  | 6 www.neu.edu/votersP R O P R I E T A R Y 

Page 7: VOTERS - Northeastern University · • Automatic quantitative pavement surface analysis, indicates pavement health (surface profile and crack density/porosity), ice, water, oil cover.

VOTERS ‐ YEAR 1VOTERS  YEAR 1

• Theme: Start R&D and Engineering on allTheme: Start R&D and Engineering on all individual VOTERS components (except MAP)

– Significant technical progress in many areasg p g y– Specifications for all subsystems are nearing completion

• Welcome Trilion Quality Systems to the VOTERS Joint Venture– In process added optical sensor system to VOTERS as important sensor subsystem

Feb. 25th, 2010  | 7 www.neu.edu/votersP R O P R I E T A R Y 

Page 8: VOTERS - Northeastern University · • Automatic quantitative pavement surface analysis, indicates pavement health (surface profile and crack density/porosity), ice, water, oil cover.

ManagementManagement

Feb. 25th, 2010  | 8 www.neu.edu/votersP R O P R I E T A R Y 

Page 9: VOTERS - Northeastern University · • Automatic quantitative pavement surface analysis, indicates pavement health (surface profile and crack density/porosity), ice, water, oil cover.

Areas of Responsibility

Management & Administrative Support Huston

Wang

Director: Ming L. Wang

Deputy Director:

TrilionBirken

Deputy Director: Sara Wadia‐Fascetti

Program Manager:J ff D ht

Trilion

Yu

Jeff Doughty

Consultants:• Infrasense

l

Rappaport

Supporting Organizations

• G. McDaniel• D. Busuioc

Trilion

Feb. 25th, 2010  | 9 www.neu.edu/votersP R O P R I E T A R Y 

Page 10: VOTERS - Northeastern University · • Automatic quantitative pavement surface analysis, indicates pavement health (surface profile and crack density/porosity), ice, water, oil cover.

Management and Evaluation ProcessesManagement and Evaluation Processes

• Regular team meetings joining JV partners andRegular team meetings joining JV partners and consultants

• Progress will also be monitored by GANTT charts

• Evaluation will be governed by decision points

Feb. 25th, 2010  | 10 www.neu.edu/votersP R O P R I E T A R Y 

Evaluation will be governed by decision points

Page 11: VOTERS - Northeastern University · • Automatic quantitative pavement surface analysis, indicates pavement health (surface profile and crack density/porosity), ice, water, oil cover.

Technical Highlightsg g

Feb. 25th, 2010  | 11 www.neu.edu/votersP R O P R I E T A R Y 

Page 12: VOTERS - Northeastern University · • Automatic quantitative pavement surface analysis, indicates pavement health (surface profile and crack density/porosity), ice, water, oil cover.

VOTERS SYSTEM DIAGRAMVOTERS SYSTEM DIAGRAM

TEAS

GEARS

TEASe

RADAR Materials Characterization

DataRegistration

OnboardProcessing

MAP

BOSSCharacterization

SLiMR

SupportingSensor Systems

Processing

SOPRA

Feb. 25th, 2010  | 12 www.neu.edu/votersP R O P R I E T A R Y 

Page 13: VOTERS - Northeastern University · • Automatic quantitative pavement surface analysis, indicates pavement health (surface profile and crack density/porosity), ice, water, oil cover.

Sensing Technologies TEASe: Tire Excited Acoustic Sensing• Adaptation of SONAR concepts to road surface analysis. Assesses subsurface

conditions.conditions.

GEARS: Gigahertz Electromagnetic Array Roaming Sensor • Radar system built to operate under vehicles at driving speed for on-the-fly

i i A b f di iinterpretation. Assesses subsurface conditions.

SLiMR: Surface Looking Millimeter Wave Radar• Automatic quantitative pavement surface analysis indicates pavement health• Automatic quantitative pavement surface analysis, indicates pavement health

(surface profile and crack density/porosity), ice, water, oil cover.

SOPRA: Surface Optical Profilometry Roadway Analysis• Phase profilometry technology uses digital pictures illuminated by periodic

shadows to profile road surfaces. Map roadway and bridge deck surface andobserve active cracking.

Feb. 25th, 2010  | 13 www.neu.edu/votersP R O P R I E T A R Y 

Page 14: VOTERS - Northeastern University · • Automatic quantitative pavement surface analysis, indicates pavement health (surface profile and crack density/porosity), ice, water, oil cover.

Design Goals ClarifiedDesign Goals ClarifiedDetecting VOTERS Sensor Subsystem

Defect Category Primary Secondary Tertiary

ace Surface Contour  SOPRA TEASe SLiMR

Surface Texture SOPRA/TEASe SLiMR

Surf

Surface Texture SOPRA/TEASe SLiMRSurface Condition(Materials Covering Roadway) SLiMR SOPRADebonding TEASe ‐ GEARSConcrete Subsurface TEASe GEARS

ubsurface

Concrete Subsurface TEASe GEARS ‐Rebar and Corrosion GEARS Lens GEARS ‐Roadway Layers GEARS ‐ TEASeSubsurface Moisture GEARS ‐ ‐

Su Expansion Joint Condition TEASe GEARS ‐Crack Activity SOPRA ‐ ‐

Density ‐ TEASe ‐

Feb. 25th, 2010  | 14 www.neu.edu/votersP R O P R I E T A R Y 

Page 15: VOTERS - Northeastern University · • Automatic quantitative pavement surface analysis, indicates pavement health (surface profile and crack density/porosity), ice, water, oil cover.

Measured Acoustic Wave from Road Vibrations

4

5Microphone time history

6

8Acceleration time history

Pressure peak predicted from plane wave acoustic model

and acceleration peak

1

2

3

re (P

a)re (P

a)

0

2

4

ion

(g)

on (g)

-2

-1

0

Air

pres

suPressur

-8

-6

-4

-2

Acc

eler

at

Accelerometer is 0.5 m from the impact

Accelerati

-0.5 0 0.5 1 1.5 2 2.5

x 10-3

-4

-3

Time (s)Time (s)

-0.5 0 0.5 1 1.5 2 2.5

x 10-3

-12

-10

Time (s)

impact

Time (s)

Long Rayleigh wavelength

Feb. 25th, 2010  | 15 www.neu.edu/votersP R O P R I E T A R Y 

Page 16: VOTERS - Northeastern University · • Automatic quantitative pavement surface analysis, indicates pavement health (surface profile and crack density/porosity), ice, water, oil cover.

Acoustic Sensor for Subsurface DamageAcoustic Sensor for Subsurface Damage

• Robust acoustic signature  • Wireless sensor design for gdetected for subsurface road damage

gdynamic tire pressure

microPa)

Resonance Feature ~ 200 Hz

Debonded Sensor inside tire

e (dB re 20 

Pressure

Frequency (Hz)

Feb. 25th, 2010  | 16 www.neu.edu/votersP R O P R I E T A R Y 

Page 17: VOTERS - Northeastern University · • Automatic quantitative pavement surface analysis, indicates pavement health (surface profile and crack density/porosity), ice, water, oil cover.

GEARSGEARS• Technical challenges, initial specifications,

d l ti th id tifi d

Pulse Generator Circuit Board

and solution paths identified

• Pulse Generator – Strip line units developed,including controllable pulse shaping and high‐ Measured Gaussian Pulseg p p g gfrequency up conversion, signal amplified and launched through ultra‐wideband antenna

D t t i ti Picosecond Pulsar Metal Plate vs No• Demonstrate innovative,high performance, full‐waveform signal sampling 0.005

0.01

0.015

0.02

V)Picosecond Pulsar Metal Plate vs No

Plate Under Concrete Slab subtracted

Plate

No Plate

of launched signal

-0.02

-0.015

-0.01

-0.005

00 2 4 6 8 10 12 14

Volta

ge (V

Feb. 25th, 2010  | 17 www.neu.edu/votersP R O P R I E T A R Y 

-0.025 Time (nanoseconds)

Page 18: VOTERS - Northeastern University · • Automatic quantitative pavement surface analysis, indicates pavement health (surface profile and crack density/porosity), ice, water, oil cover.

Novel Antenna Concepts for GPR  FocusingAdding a delay line: td

•Focusing Lens Rebar Detector•Low profile Vivaldi feed elementLow profile Vivaldi feed element•Compact offset parabolic mirror 

Alternative P b l id

Truck Undercarriage

Feed

Paraboloids 

Collimated Rays

Feb. 25th, 2010  | 18 www.neu.edu/votersP R O P R I E T A R Y 

Page 19: VOTERS - Northeastern University · • Automatic quantitative pavement surface analysis, indicates pavement health (surface profile and crack density/porosity), ice, water, oil cover.

Adjusted GEARS PlanAdjusted GEARS Plan

Feb. 25th, 2010  | 19 www.neu.edu/votersP R O P R I E T A R Y 

Page 20: VOTERS - Northeastern University · • Automatic quantitative pavement surface analysis, indicates pavement health (surface profile and crack density/porosity), ice, water, oil cover.

SLiMR:  Custom Built and COTS Radars• Distinguishes water‐infiltrated and Ice‐

coated roadway• Alerts on change in pavement condition• Alerts on change in pavement condition 

with changing position

T ittTransmitter

Power

RF Reflectivity of Road Surfaces

12141618

Receiver

Power

02468

10

seline

sphaltsphaltsphalttated

phaltmin cre

tecre

te

crete

crete

rete rete

mV

Tx antenna Rx antenna

Base

Dry Asp

Dry Asp

Dry Asp

Dry Asphalt r

otatWet A

sp

Wet asphalt a

fter 3

mDry

concrDry

concrWet C

oncrWet C

oncr

~6mm ice on co

ncre

~6mm ice on co

ncre

Feb. 25th, 2010  | 20 www.neu.edu/votersP R O P R I E T A R Y 

Page 21: VOTERS - Northeastern University · • Automatic quantitative pavement surface analysis, indicates pavement health (surface profile and crack density/porosity), ice, water, oil cover.

RADAR MATERIALS CHARACTERIZATION

GEARS

RADAR MATERIALS 

TEASe

DataRegistration

Onboard

MAP

BOSS

– Dielectric database: 663 data points from 45 journal papers (until Feb. 21, 2010)

E i t l t f ilit C t t d CHARACTERIZATION

SLiMR

SupportingSensor Systems

Processing

ORSPA

BOSS

SOPRA

– Experimental measurement facility: Contact and non‐contact methods for 0.1 GHz to 20 GHz

– Theoretical investigation: Dielectric mixture d l f t h d th h i t

PCC specimens

models for two‐phase and three‐phase mixtures

PCC specimens

1 GHz5 GHz20 GHz 1 GHz 5 GHz 20 GHz

w/c = 0.35 w/c = 0.40

/

w/c = 0.35 w/c = 0.40

/Asphalt specimens

1 GHz 5 GHz 20 GHzw/c = 0.42 w/c = 0.45

w/c = 0.50 w/c = 0.55

w/c = 0.42 w/c = 0.45

w/c = 0.50 w/c = 0.55

Feb. 25th, 2010  | 21 www.neu.edu/votersP R O P R I E T A R Y 

1 GHz 5 GHz 20 GHz1 GHz 5 GHz 20 GHz

Page 22: VOTERS - Northeastern University · • Automatic quantitative pavement surface analysis, indicates pavement health (surface profile and crack density/porosity), ice, water, oil cover.

SOPRA S rface Optical Profilometr Road a Anal sisSurface Optical Profilometry Roadway Analysis

• Road surface shape in 3D

• Roadway damage detection‐ Pot hole‐ Crackingg‐ Expansion Joints

• Optimized for 5mm resolution 2 iover 2m view

• Real‐time distance to surface

Vi i i t ti d t• Vision registration data

• Continuous road surface image

Feb. 25th, 2010  | 22 www.neu.edu/votersP R O P R I E T A R Y 

g

Page 23: VOTERS - Northeastern University · • Automatic quantitative pavement surface analysis, indicates pavement health (surface profile and crack density/porosity), ice, water, oil cover.

BOSS and SYSTEM INTEGRATIONMAJOR ACCOMPLISHMENTS 

• Put together experienced system integration teamg p y g

• Determined overall high‐level VOTERS system architecture

• Selected BOSS building blocks andassembled initial basic prototype

• Decided on following VOTERSstrategies:strategies:– 3‐Level Decision Making

– GPS based Timing Synchronization

– Data Collection Mode

• Working on Data Registration strategy

Feb. 25th, 2010  | 23 www.neu.edu/votersP R O P R I E T A R Y 

Page 24: VOTERS - Northeastern University · • Automatic quantitative pavement surface analysis, indicates pavement health (surface profile and crack density/porosity), ice, water, oil cover.

DGPSB f i tBase:  reference point

Rover: updates its accurate position

• Develop software from scratch.

I th t h lf t i i i• Improve the accuracy to half‐meter using inexpensivereceiver (<$100).

Trimble Copernicus GPS chip ($80)

Compact Magnetic‐Mount  Antenna ($10)

• Communicate via commercial wireless channel.

($80)

Feb. 25th, 2010  | 24 www.neu.edu/votersP R O P R I E T A R Y 

Page 25: VOTERS - Northeastern University · • Automatic quantitative pavement surface analysis, indicates pavement health (surface profile and crack density/porosity), ice, water, oil cover.

DGPS Results0 20 40 60 80 100 120 140 160 180 200

-1

0

1

Eas

t erro

r [m

]

DGPS on position: Rover1 error componets (east, north and vertical)

Point ABase Station

Time[s]

0 20 40 60 80 100 120 140 160 180 200

0

2

4

Time[s]

Nor

th e

rror [

m]

Point B

Point BRover 1

Time[s]

0 20 40 60 80 100 120 140 160 180 200-2

0

2

Time[s]

Ver

tical

erro

r [m

]

0 20 40 60 80 100 120 140 160 180 200

0

1

2

Time[s]

Eas

t erro

r [m

]

DGPS on position: Rover2 error componets (east, north and vertical)

Point CRover 2

Time[s]

0 20 40 60 80 100 120 140 160 180 200-2

0

2

Time[s]

Nor

th e

rror [

m]

Accuracy

Point C

[ ]

0 20 40 60 80 100 120 140 160 180 200-4

-2

0

Time[s]

Ver

tical

erro

r [m

]

AccuracyPoint B: East: 1 meters, North: 2 meters,      

Vertical 0.2 meters.

Point C: East: 1 meters, North: 0.5 meters, 

Vertical 2 meters

Feb. 25th, 2010  | 25 www.neu.edu/votersP R O P R I E T A R Y 

Vertical 2 meters.

Page 26: VOTERS - Northeastern University · • Automatic quantitative pavement surface analysis, indicates pavement health (surface profile and crack density/porosity), ice, water, oil cover.

Outreach summaryOutreach summary

• Patent disclosures: 4Patent disclosures: 4

• Papers: 5

C f b i i• Conference submissions: 5

• Proprietary issue is hold back publications

• Public relation: magazines and newspapers

• Website: www neu edu/votersWebsite: www.neu.edu/voters

• Industry interest : 20

Feb. 25th, 2010  | 26 www.neu.edu/votersP R O P R I E T A R Y 

Page 27: VOTERS - Northeastern University · • Automatic quantitative pavement surface analysis, indicates pavement health (surface profile and crack density/porosity), ice, water, oil cover.

Upcoming VOTERS Year 2Upcoming VOTERS Year 2

• Theme: Complete experimental prototypes ofTheme: Complete experimental prototypes of individual VOTERS components, start integration and field testingintegration and field testing

• First sensor subsystem experimental prototypes will become available (e g SLiMRprototypes will become available (e.g. SLiMR, Sept. 2010)

E j d i TEAS GEARS• Expect major advances in TEASe, GEARS,  SOPRA, and DGPS

Feb. 25th, 2010  | 27 www.neu.edu/votersP R O P R I E T A R Y 

Page 28: VOTERS - Northeastern University · • Automatic quantitative pavement surface analysis, indicates pavement health (surface profile and crack density/porosity), ice, water, oil cover.

Expanding VOTERSExpanding VOTERS

• Other roaming sensor ideasOther roaming sensor ideas

• Other interest: Airports

ild i d i• Build on industry interest

• Link MAP into existing roadway and bridge deck management systems and demonstrate added value

Feb. 25th, 2010  | 28 www.neu.edu/votersP R O P R I E T A R Y 

Page 29: VOTERS - Northeastern University · • Automatic quantitative pavement surface analysis, indicates pavement health (surface profile and crack density/porosity), ice, water, oil cover.

SummaryyWhat do we like to achieve today ?

• Team work

f• Report to NIST of our progress and new opportunity

• Seek guidance from NIST TIP• Seek guidance from NIST‐TIP program manager and the team

Feb. 25th, 2010  | 29 www.neu.edu/votersP R O P R I E T A R Y