www.ptvgroup.comwww.ptvgroup.com

DYNAMIC ODME FOR AUTOMATED VEHICLES

MODELING USING ‘BIG DATA’

Dr. Jaume Barceló, Professor Emeritus, UPC-

Barcelona Tech, Strategic Advisor to PTV Group

Shaleen Srivastava, Vice-President/Regional

Director (PTV Group Americas)

www.ptvgroup.com

INTRODUCTORY REMARKS

Connected vehicle systems and autonomous vehicles likely to be major

game changers in traffic and mobility. No longer a question of if, but of

when, in what form, at what rate. And through what kind of evolution

path…

… operational regimes in which vehicles are connected to each other

and to the infrastructure, and augmented with autonomous capabilities.

(Hani Mahmassani, Workshop 134: Emerging Needs for Improving Simulation Models, TRB 96th

Annual Meeting, Washington, January 8, 2017)

www.ptvgroup.com

CONNECTED & AUTONOMOUS VEHICLES

www.ptvgroup.com

AUTONOMOUS VS CONNECTED VEHICLES

www.ptvgroup.com

SINGLE VEHICLE APPLICATIONS &

COOPERATIVE APPLICATIONS

www.ptvgroup.com

WILL AV REPLACE CURRENT AUTOMOTIVE

TECHNOLOGIES FOR INDIVIDUAL MOTORIZED MOBILITY?

OR, WILL MOSTLY BE USED FOR COLLECTIVE MOBILITY?

http://www.traffictechnologytoday.com/video-audio.php?v=GATEwayTeam

www.ptvgroup.com

VISUALIZATION OF SHARED SELF-DRIVING CAR

SIMULATION FOR LISBON

www.ptvgroup.com

A SELF-ORGANIZING SYSTEM OR BETTER

EXTERNALLY ASSISTED?

www.ptvgroup.com

COOPERATIVE DRIVING WITH THE HELP OF V2X COMMUNICATIONS

Source: D. Jia, D. Ngoduya,

Enhanced cooperative car-

following traffic model with the

combination of V2V and V2I

communication

Transp. Res. B, March 2016

Source: L. Zhao, J. Sun, Simulation Framework for Vehicle Platooning and

Car-following Behaviors under Connected-Vehicle Environment, Procedia -

Social and Behavioral Sciences 96 ( 2013 ) 914 – 924.

www.ptvgroup.com

TESTING VACS BY MICROSOPIC SIMULATION

From: Ntousakis, I.A., Nikolos, I.K., Papageorgiou, M.: On microscopic modelling of adaptive cruise control systems. 4th Intern.

Symposium of Transport Simulation (ISTS’14), 1-4 June 2014, Corsica, France. Transportation Research Procedia 6 (2015), pp. 111-127.

ACC string-stability ACC traffic efficiency

www.ptvgroup.com

ACC/CACC: STABILITY/EFFICIENCY

Macroscopic simulation of traffic flow (spatio-temporal evolution of traffic density) close to an on-ramp using

the GKT model, combined with a novel ACC/CACC modeling approach. Left: manual cars; Middle: ACC-

equipped cars; Right: CACC-equipped cars.

Source: Delis, A.I., Nikolos, I.K., Papageorgiou, M.: Macroscopic traffic flow modeling with adaptive cruise

control: development and numerical solution. Computers & Mathematics with Applications, 2015

www.ptvgroup.com

HIERARCHICAL+ TM

Network Traffic Control

Link Control Link Control

V2I

V2V

• Overlapping link controllers?

• Share of control tasks?

• Connect VACS and TM communities for

maximum synergy

• TM remains vital while VACS are emerging

Papageorgiou, M., Diakaki, C., Nikolos, I., Ntousakis, I., Papamichail, I., Roncoli, C. : Freeway traffic management in

presence of vehicle automation and communication systems (VACS). In Road Vehicle

Automation 2, G. Meyer and S. Belker, Editors, Springer International Publishing, Switzerland, 2015, pp. 205-214

www.ptvgroup.com

EQUIPPED VEHICLES, V2V & V2I BECOME RICH DATA

SOURCES TO SUPPORT MANAGEMENT AND GUIDANCE

Autonomous vehicles rely on knowing the roadway they are

traveling on, changes to the roadside such as

new development or construction will require the type of

real-time exchange of information that CV technology

provides including valuable information about the road

ahead—allowing rerouting based on new information such

as a lane closures, or congestion growing.

ATKINS “Autonomous vs connected vehicles – what’s the difference?” (Suzanne

Murtha | 02 Oct 2015 |)

www.ptvgroup.com

SENSORS, FUNCTIONS AND SURROUNDING “AWARENESS” OF PROBE

CARS – DEPICTED IN RED- 4 VEHICLES DETECTED IN FRONT AND 3 BEHIND

X-X

-Y

Y

Each of the vehicles measures each

half second and stores its position,

heading, speed, and acceleration,

as well as distances and relative

speeds to “visible” surrounding

vehicles captured by the radars.

www.ptvgroup.com

f ldfl

(xf, yf) (xl, yl)(a)

dfl

(xf, yf) (xl, yl)

f l

(b)

f deo

(xe, ye)

(xo, yo)

o(c)

A

B

1

2 3

4

5

6

7

8V2V

GPS Equipped

Bluetooth Equipped

Unequipped

Traffic Data Center2A5

1

34

Time t

Space x

VW equipped

A

B

1

2 3

4

5

6

7

8V2V

GPS Equipped

Bluetooth Equipped

Unequipped

Traffic Data Center2A5

1

34

Time t

Space x

VW equipped

V2V TRACKED EQUIPPED VEHICLES “AWARE” OF

SURROUNDING VEHICLES TRAJECTORY

RECONSTRUCTION & TRAFFIC STATE ESTIMATION

•The relative distance deo

between the equipped

and the observed car

•The relative speed veo

between the equipped

and the observed car

•The map-matched

position (xe,ye) and speed

ve of the equipped car. Source: L. Montero, J. Barceló et al., A case study on cooperative car data

for traffic state estimation in an urban network, Paper 16-4959, 95th TRB

Annual Meeting 2016, Compendium of Papers.

www.ptvgroup.com

(Extracting the most useful & valuable information from traffic measurements)

Dealing with heterogeneous traffic data from varied technological sources (conventional detectors,

Bluetooth, GPS, cooperative & autonomous vehicles…):

- Data filtering, completion and fusion techniques

- Processing huge amounts of data (Big Data Ad hoc Data Base Management Techniques)

Kernel Smoothing Methods,

Kalman Filter & traffic flow

based models to identify and

remove outliers

And to supply missing data

Kernel Smoothing

Methods

Machine Learning

Traffic Models

Dynamic Flow Models

OD Estimation

Filtering and

completion

techniques

Data

Fusion

Techniques

TRAFFIC DATA ANALYTICS

www.ptvgroup.com

CONCEPTUAL APPROACH TO AN ADAPTIVE AREA WIDE

CONTROL STRATEGY BASED ON THE NETWORK FLOW DIAGRAM

Figure 6 Potential use of the Network Fundamental Diagram to support Active Area WideTraffic Management

Strategies

URBAN AREA TO MANAGE

LARGE URBAN OR METROPOLITAN AREA

Origin r

Destination s

Congestion

Alternative recommended

route

GATE-OUT

GATE-IN

QUEUE

Estimation algorithm for 𝒏 𝒌 ADAPTIVE FLOW CONTROL STRATEGY

A

B

Critical Point in the managed area

Allow access Restrict access

C

Real-time Traffic Data

Measurements from sensors Output flows

n(k-1)

Input flow

rates (k)

M. Keyvan-Ekbatani, M. Papageorgiou,

V. L. Knoop, Comparison of On-Line Time-Delayed

and Non-Time-Delayed Urban Traffic Control via

Remote Gating, TRB 2015 Annual Meeting, Paper

15-4289

www.ptvgroup.com

TRAFFIC DATA ANALYTICS: DYNAMIC OD ESTIMATION

( )−++++++ += k

T

1k

k

1k1k

T

1k

k

1k1k RFPFFPG

( )( )k

1k1k1k1k gFGd ++++ −+= 1z k

( ) k

1k1k1k

1k

1k PFGIP +++

+

+ −=

k

Tk

k

k

1k WDDPP +=+k

k

k

1k Dgg =+Initialization

KF recursive

dynamics

01 += ++

+

+ 1k

k

k

1k

1k dgg

Ad Hoc Kalman Filter to estimate the time dependent OD

J.Barceló, L.Montero, M.Bullejos, M.P. Linares, O. Serch (2013), Robustness and computational efficiency of a Kalman Filter estimator of time dependent OD matrices exploiting

ICT traffic measurements. TRR Transportation Research Records: Journal of the Transportation Research Board, No. 2344, pp. 31-39.

State equations AR(r) on deviates:

Observation equations:

( ) ( ) ( ) ( )kw1wkgwD1kg1

++−=+ =

r

w

D(w) transition matrices describing the

effects of previous OD path flow deviates

gijc(k-w+1) on current flows gijc(k+1)

First block: deviates of observations at sensor locations

Second block: conservation flows for each time interval k

( )

(k)v

(k)v

(k)v

kz

3

2

1

22

11

v(k)g(k)F(k)g(k)

E(k)

(k)UA

(k)UAT

T

+=

+

=

Origin

Destination

τpijt number of trips from Origin i to

Destination j in time period for purpose p

Identification of time-dependent mobility

patterns in terms of Origin-Destination (OD)

Matrices Exploiting ICT measurements

MLU OD

Path id

OD path

links

OD

pair

ICT

sensor id

Entry

id

MLU

OD

Path id

OD path

links

OD

pair

ICT

sensor

id

Entry

id

1 1-6-11 1=(1,8) 6, 1,5 1 6 3-6-11 3=(2,8) 7,1,5 2

2 2-7-9-11 1=(1,8) 6,2,3,5 1 7 4-7-9-11 3=(2,8) 7,2,3,5 2

3 2-8-13 1=(1,8) 6,2,4 1 8 4-8-13 3=(2,8) 7,2,4 2

4 1-5-10-14 2=(1,9) 6,1,3,4 1 9 3-5-10-14 4=(2,9) 7,1,3,4 2

5 2-8-14 2=(1,9) 6,2,4 1 10 4-8-14 4=(2,9) 7,2,4 2

www.ptvgroup.com

DATA COLLECTION FROM AUTONOMOUS/CONNECTED VEHICLES

Assumption: travel times Trq of drivers

departing from origin r during time

interval t going through POI q follow a

distribution (not stationary under

congestion), no matter the selected path.

Approximate travel time distributions

by discrete distributions with bin

proportions updated according to

collected on-line ICT data.

www.ptvgroup.com

EKF APPROACH FOR NETWORKS (III) : FLOW

ESTIMATES AND ERROR CORRECTIONS (SHALEEN

SRIVASTAVA, 2010)

• Traffic flow at a location

• Flows – y(t)

• Assume Gaussian distributed measurements

• Model simulation (virtual detectors) – traffic flow

• Measurement (real detectors) – traffic flow

• Flows measurement from the model at t1:

Mean = z1

Variance = σz1

• Optimal estimate of traffic flows: ŷ(t1) = z1

• Variance of error in estimate: σ2x (t1) = σ2z1

www.ptvgroup.com

EKF APPROACH FOR NETWORKS (III) : FLOW

ESTIMATES AND ERROR CORRECTIONS (SHALEEN

SRIVASTAVA, 2010)

• So we have the prediction ŷ-(t2)

• Detector data measurement at t2: Mean

= z2 and Variance = σz2

• Need to correct the prediction by model

due to measurement to get ŷ(t2)

• Closer to more trusted measurement –

linear interpolation

• Corrected mean is the new optimal

estimate of traffic flows (basically we

have ‘updated’ the predicted flows by

model using detector data)

• New variance is smaller than either of

the previous two variances

www.ptvgroup.com

EKF APPROACH FOR NETWORKS (III) : FLOW

ESTIMATES AND ERROR CORRECTIONS (SHALEEN

SRIVASTAVA, 2010)

• If measurement is preferred:

- Measurement error covariance decreases to zero

- Weights residual more heavily than prediction

• If prediction is preferred:

- Prediction error covariance decreases to zero

- Weights prediction more heavily than residual

www.ptvgroup.com

MEASURING THE QUALITY OF THE ESTIMATES ESTIMATED (KF

APPROACH) VS TARGET FLOWS IN OD PAIRS FOR A 15 MINUTES

INTERVAL

0

50

100

150

200

250

300

350

400

450

500

1 4 6 7 8 14 19 21 22 30 36 43 44 45 55 56 58 61 68 70 76 77 85 2 9 10 15 16 17 24 27 29 31 34 35 38 50 52 54 71 72 75 78 80 3 5 11 12 13 23 26 37 40 41 48 49 53 57 63 64 66 67 73 82 18 20 25 28 32 33 39 42 46 47 51 59 60 62 65 69 74 79 81 83 84

Esti

mat

ed

vs

Targ

et

OD

Flo

ws

-1

h (

veh

/h

OD Pair

Target OD flow

Estimate OD flow

y = 1,04x - 2,9644R² = 0,9387

0

50

100

150

200

250

300

350

400

450

500

0 50 100 150 200 250 300 350 400 450

Demand Set 2: Estimated vs Target OD flows - 1h

Barcelona’s Central

Business District

(CBD), Eixample, 2111

sections, 1227 nodes

120 generation

centroids, 130

destination centroids

(877 non-zero OD pairs)

116 Loop detector

Stations & 50 Bluetooth

Antennas

www.ptvgroup.comwww.ptvgroup.com