DYNAMIC ODME FOR AUTOMATED VEHICLES … PTV Presentation SF... · DYNAMIC ODME FOR AUTOMATED...
Transcript of DYNAMIC ODME FOR AUTOMATED VEHICLES … PTV Presentation SF... · DYNAMIC ODME FOR AUTOMATED...
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