SlideShare a Scribd company logo
1 of 42
Intelligent Traffic Light Control
Mohamed Khasawneh
CIISE, Concordia University
PhD Comprehensive Exam
1
Outline
1. Introduction
2. Context
3. Papers
1. Intelligent Traffic Light Control Using Distributed Multi-agent Q Learning
2. Intelligent Traffic Lights: Green Time Period Negotiation
3. Optimal Cycle Program of Traffic Lights with Particle Swarm Optimization
4. Conclusions
5. Timeline
6. Future works
2
1. Introduction
• Nowadays, studying Intelligent Transport Systems (ITS) is considered to be one of the hottest
topics which can improve the quality of life in many ways. Designing a good ITS system can bring
us many benefits such as saving lives by reducing the percentage of accidents, increasing the
safety for pedestrians, reducing the waiting times for both vehicles and pedestrians, optimizing
fuel consumption, reducing the emissions of CO and CO2 gases, reducing the noise pollution, etc.
3
2. Context
• Smart cities
• Traffic Congestion management problem
• Intelligent Traffic signaling problem
• Connected vehicles
4
Challenges
• There are many challenges in designing these systems:
The lack of global view of traffic conditions at the neighboring streets or intersections.
The unexpected change in traffic condition at any given time.
The high congestion status which makes it a hard and daunting task to accomplish
optimization between vehicles and pedestrian in a fair and satisfying manner.
5
Papers under study
1. Intelligent Traffic Light Control Using Distributed Multi-agent Q Learning.
2. Intelligent Traffic Lights: Green Time Period Negotiation.
3. Optimal Cycle Program of Traffic Lights with Particle Swarm Optimization.
6
Paper 1.
Intelligent Traffic Light Control Using Distributed
Multi-agent Q Learning
7
Problem Definition
• Designing Intelligent traffic light control, where two main aspects are not considered in the
previous studies:
Non motorized traffic.
Global optimization.
8
Literature Review
9
Figure 1.Classification Of Traffic Light Control Approach
Proposed System Design
• The proposed system design is based on the usage of Q learning which falls under the
reinforcement learning techniques.
• Q learning technique is composed of two main phases which are exploration and exploitation
phases.
• In exploration phase, the Q learning agent is randomly exploring the environment around, after
the environment is learned an exploitation phase is starting with an objective to maximize the
rewards in the whole system.
10
Proposed System Design
Figure 2. System Architecture for Distributed Q Learning Agent [2].
11
Methodology
• A Markov Decision Based model where states, actions, rewards are defined.
States :
𝑆𝑖,𝑑
𝑡
= {𝑞1𝑖,𝑑
𝑡
, 𝑞2𝑖,𝑑
𝑡
, … , 𝑞 𝑗𝑖,𝑑
𝑡
, 𝑚1𝑖,𝑑,𝐿
𝑡
, 𝑚1𝑖,𝑑,
𝑡
, … , 𝑚1𝑖,𝑑
𝑡
, 𝑚1𝑖,𝑑
𝑡
} (1)
Actions:
Fig. 3: Action sets (green lights) for “+” shape intersection
12
Methodology
Rewards:
𝑹𝒊,𝒅
𝒕
𝒂𝒊,𝒅
𝒕
, 𝒂𝒋,𝒅
𝒕
, 𝑺𝒊,𝒅
𝒕
, 𝑺𝒋,𝒅,
𝒕
𝑾𝒊,𝒅
𝒕
= −(
𝑾 𝟏,𝒅
𝒕
𝑵 𝒊
𝒋𝜺𝑵 𝒊
𝒒𝒋𝒊,𝒅
𝒕
+
𝑾 𝒛,𝒅
𝒕
𝑵 𝒊 𝑵 𝒋
𝒋𝜺𝑵 𝒊 𝒌𝜺𝑵 𝒋
𝒒 𝒌𝒋,𝒅
𝒕
+
𝑾 𝟑,𝒅
𝒕
𝟐 𝑵 𝒊
𝒋𝜺𝑵 𝒊
( 𝒎𝒋𝒊,𝒅,𝑳
𝒕
+ 𝒎𝒋𝒊,𝒅,𝑹
𝒕
)) (2)
Integrate real-world constraints and traffic rules.
13
Used parameters
Parameters Meaning
𝑺𝒊,𝒅
𝒕 The state at intersection i, at day d and time t.
𝒒 𝟏𝒊,𝒅
𝒕
, 𝒒 𝟐𝒊,𝒅
𝒕
, … , 𝒒𝒋𝒊,𝒅
𝒕 The queue length from intersection j to intersection i, at
day d and time t
𝒎 𝟏𝒊,𝒅,𝑳
𝒕
, 𝒎 𝟏𝒊,𝒅,𝑹
𝒕 The queue length for pedestrians at the left side from
intersection j to i, at day d and time t,
𝑹𝒊,𝒅
𝒕 The reward at intersection i, at day d and time
𝑾 𝟏,𝒅
𝒕 The weight to present the local vehicular queues at
intersection I
𝑾 𝟐,𝒅
𝒕 The weight to present the neighborhood vehicular
queues at the neighbors of intersection I
𝑾 𝟑,𝒅
𝒕 Weight to present the total pedestrian queues at
intersection I
𝒂𝒊,𝒅
𝒕 The action at intersection i, at day d and time t
14Table 1. Used Parameters
Results
Fig. 4: Total cumulated queue lengths for vehicles and
pedestrians in the case of high pedestrian rate
Fig. 5: Total waiting and traveling time for vehicles and
pedestrians in the case of high pedestrian rate
15
Results
Fig. 6: CO2 and CO emission of vehicles in the case
of high pedestrian rate
Fig. 7: Fuel consumption and noise pollution of vehicles in the
case of high pedestrian rate
Fig. 8: Total cumulated queue lengths for vehicles and
pedestrians in the case of medium pedestrian rate
Fig. 9: Total waiting and traveling time for vehicles and
pedestrians in the case of medium pedestrian rate
16
Conclusion
• Unlike other papers in the literature, one of the main contributions of this paper is handling both
motorized and non-motorized vehicles. Non-motorized traffic such as pedestrian and bicyclist are
also an important part of the whole system and should not be overlooked in order to achieve the
fairness among all users. Other contribution is the achievement of near global optimal solution by
gathering/exchanging all the information from/to local and neighboring intersections, this helps
in a better and more accurate decisions compared with other proposed algorithms in literature
which are either using fixed-time or dynamic control approach
17
Paper 2:
Intelligent Traffic Lights: Green Time Period
Negotiation
18
Problem Definition
• Designing Intelligent traffic light control, where two main aspects are not considered in the
previous studies:
Person-based.
The distribution of green time period between the traffic streams.
19
Literature Review
20
Author Approach
Dresner and Stone (2005) (2009) “reservation based” strategy
Vasirani and Ossowski(2009) (2011) “market based” strategy and extended
the work done by Dresner and Stone
Schepperle and Bohm(2008) this strategy will take in account
valuation of the driver
Table 2. Literature Review
Proposed System Design
• A multi-agent computerized system (MAS or "self-organized system") composed of multiple
interacting intelligent agents is proposed. Two stages are involved:
First stage : try to find a traffic plan through defining the phase and determining green time
periods
Second stage: include two decisions defining the time of terminating the current stage and
defining the next stage to be implemented.
21
First stage
𝑵 𝟎 =
𝑸𝑻 𝒇
𝟒
(𝑿𝒊−𝟏) + 𝒙𝒊 − 𝟏 𝟐 +
𝟏𝟐 𝒙 𝒊− .𝟔𝟕+
𝒔𝒈
𝟔𝟎𝟎
𝑸𝑻 𝒇
𝒊𝒇 𝑿𝒊 >. 𝟔𝟕 +
𝒔𝒈
𝟔𝟎𝟎
(3)
0 otherwise
𝑵 𝑴𝒂𝒙 =
𝒒 𝑪−𝒈
𝟏−𝒚
+ 𝑵 𝟎
𝒏𝒖𝒎𝒃𝒆𝒓 𝒐𝒇 𝒍𝒂𝒏𝒆𝒔
< 𝒍𝒂𝒏𝒆 𝒄𝒂𝒑𝒂𝒄𝒊𝒕𝒚 𝒒𝒖𝒆𝒖𝒆 (4)
𝑫 𝒕𝒐𝒕𝒂𝒍=𝒗𝒊𝒆𝒄𝒉𝒍𝒆𝒔+𝒑𝒆𝒅𝒆𝒔𝒕𝒓𝒂𝒊𝒏𝒔 = 𝒊=𝟏
𝑰 𝒒𝑪(𝟏−𝒈𝒊/𝑪) 𝟐
𝟐(𝟏−𝒙 𝒊 (𝒈𝒊/𝑪)
+ 𝑵 𝟎 𝑿𝒊 𝒗𝒆𝒉 𝒐𝒄𝒄 (5)
22
Second stage
23
Bid Mechanism
24
Case study
Variables values
Simulation Time 2 hours
The inter-green time 5s (3s yellow, 2s red).
The minimum green duration 7 second
The maximum planned cycle length 120 second
the maximum acceptable degree of saturation 80%
The saturation flows for straight-ahead
movements, right-turn movement and left-turn
movements
1800, 1600 and 1700 veh/h/lane, respectively.
average vehicle occupancy 1.2
Simulation Software AIMSUN
Table 3. Simulation Parameters
25
Result
Fig. 5 – Average travel time of each replication versus time: a) baseline; b) proposed
26
Conclusions
• This paper unveiled a person-oriented traffic signal control method for secluded intersection that
include signal timing optimization and concurrent signal plan designs with actuated-time data
about the dynamics of the network.
• The suggested control mechanism operates within two phases: first, it chooses traffic signals and
second, it undertakes negotiation of green time duration based on actual traffic situations. The
suggested control is assessed using a tiny traffic simulation tool.
27
Paper 3:
Optimal Cycle Program of Traffic Lights With Particle Swarm
Optimization
28
Problem definition
• Designing Intelligent traffic light control, where three main aspects are not considered in the
previous studies:
Large vehicular networks.
Many scenarios.
Comparison with other algorithms.
29
Literature Review
Mathematic models Fuzzy logic approaches Biologically inspired optimizers
McCrea and Moutari[5]( merged
knowledge-founded and continuous
calculus-founded models with the
intent of defining the road networks’
traffic flow)
 Tolba et al.( offered a Petri net-
based model to indicate traffic flow,
from both macroscopic and
macroscopic perceptions,
consideration of vehicles’ individual
trajectories and observation of
global variables, respectively
Lim et al. [4]( derived a fuzzy logic
controller during a single
intersection’s real-time local
optimization)
 Other experts [3] developed a traffic
simulator utilizing fuzzy logic agents
for traffic lights at isolated
intersections.
Cellular automata:
 Brockfeld et al. [6] utilized a CA
model whereby the city network
was instigated as a simple square
having some normal streets and
four junctions.
Neural networks (NN):
 Spall and Chin [1] illustrated an NN
considered in configuring traffic
lights’ control parameters.
Genetics Algorithm
30
Table 4.Literature Review(Paper 2)
Solution Approach
• PSO has been considered in [7] and [8] as a population-based meta-heuristic motivated by the
birds’ social behavior within a flock, whose initial purpose was to continually optimize problems.
• In this model, initial solutions are randomly generated .each solutions to the problem is
identified as particle locations while the particles’ population is known as swarm. For this
algorithm, xi, the position of each particle, is updated for each iteration g as follows:
where term 𝑣 𝑔+1
𝑖
is the velocity of the particle, given by:
31
𝑋 𝑔+1
𝑖
= 𝑋 𝑔
𝑖 + 𝑣 𝑔+1
𝑖
𝑣 𝑔+1
𝑖
= 𝑤. 𝑣𝑔
𝑖
+U[0,𝜑1]. (𝑃𝑔
𝑖
− 𝑥 𝑔
𝑖
)+U[0,𝜑2]. 𝑏 𝑔 − 𝑥 𝑔
𝑖
.
SUMO
32
PSO for Traffic Light Scheduling
The optimization solver proposed in this paper for optimal cycle programs for traffic light involves three
steps which are :
Solution Encoding: all traffic light logics with their phase durations are mapped into each solution and
those solutions are randomly generated by PSO
The fitness function :Each solution is evaluated, taking into account the information obtained from
events that occur during simulation by the following equation:
F(s)= 𝑣=0
𝑉 𝑗 𝑣 𝑠 + 𝑣=0
𝑉+𝐶 𝑤 𝑣 𝑠 +(𝐶 𝑠 .𝑆𝑡)
𝑉2 𝑠 +𝐶𝑟
The global optimization procedure: the optimization strategy is composed into two parts :
an optimization algorithm
a simulation procedure
33
Case study
34
Table 5. Simulation and PSO Parameters
Results
35
Papers comparison
Weakness Strengths
Paper 1 o Vehicle-based strategy
o does not react well to a sharp
changes in traffic patterns
o Global optimization
o Motorized and non-motorized traffic
Paper 2 o The proposed mechanism is
only admissible where the
demand of pedestrians is low.
o Person-based strategy
o the system is flexible
Paper 3 o priority for motorized traffic. o unlimited vehicular networks
o various scenario topologies
o Compared with four algorithms
36
Table 6. Papers comparison
Conclusions
• In this report, research papers tackling the efficient design of intelligent traffic light are discussed.
Considering different parameters to reach a different objective function.
• In [2], both motorized and non-motorized vehicles are taken into account to minimize the overall
congestion and waiting time. Whereas, in [9] an auction based intersection control strategy is
proposed with the objective function of minimizing the waiting time for each person by taking
into consideration the number of persons in each car.
• On the other hand, in another paper used an optimization method called Particle Swarm
Optimization (PSO) to find successful traffic light cycle programs with two main objectives which
are the number of vehicles arriving at the destination and the total travel time.
37
Timeline
38
Future works
• Designing a person based globally optimal intelligent traffic light which fairly considers many
parameters (motorized and non-motorized traffic).
• Designing a machine learning-based algorithm to predict traffic congestion and signal length
timings based on historical information
39
References
[1] Y. Feng, K. L. Head, S. Khoshmagham, and M. Zamanipour, ―A real-time adaptive signal control
in a connected vehicle environment,Transportation Research Part C: Emerging Technologies, vol. 55,
pp. 460–473, 2015.
[2] Ying Liu1, 2, Lei Liu1, Wei-Peng Chen1,Intelligent Traffic Light Control Using Distributed Multi-
Agent Q Learning, Fujitsu Laboratories of America, Inc., Sunnyvale, CA, USA
[3] Timóteo, I., Araújo, M., Rossetti, R. & Oliveira, E. 2010. TraSMAPI: An API oriented towards Multi-
Agent Systems real-time interaction with multiple Traffic Simulators. In: 13th Intern. IEEE
Conference on Intelligent Transportation Systems (ITSC), Funchal, pp. 1183-1188.
[4] Dion, F., Rakha, H. & Kang, Y.-S. 2004. Comparison of delay estimates at under-saturated and
over-saturated pre-timed signalized intersections. Transportation Research Part B: Methodological,
38(2):99-122.
[5] J. McCrea and S. Moutari, ―A hybrid macroscopic-based model for traffic flow in road
networks,Eur. J. Oper. Res., vol. 207, no. 1, pp. 676–684, 2010.
40
References
[6] R. Bretherton, N. Hounsell, and B. Radia. (1996). Public transport priority in SCOOT [Online].
Available: http://eprints.soton.ac.uk/75299/
[7] M. A. M. de Oca, T. St¨utzle, M. Birattari, and M. Dorigo, ―Frankenstein‘s PSO: A composite
particle swarm optimization algorithm,IEEE Trans. Evol. Comput., vol. 13, no. 5, pp. 1120–1132, Oct.
2009.
[8] J. Kennedy and R. C. Eberhart, Swarm Intelligence. San Francisco, CA, USA: Morgan Kaufmann,
2001.
[9] Cristina Vilarinho, José Pedro Tavares, Rosaldo J. F. Rossetti, Intelligent Traffic Lights: Green Time
Period Negotiato.
41
Thank you
42

More Related Content

What's hot

IRJET- Road Recognition from Remote Sensing Imagery using Machine Learning
IRJET- Road Recognition from Remote Sensing Imagery using Machine LearningIRJET- Road Recognition from Remote Sensing Imagery using Machine Learning
IRJET- Road Recognition from Remote Sensing Imagery using Machine LearningIRJET Journal
 
tAnt colony optimization for
tAnt colony optimization fortAnt colony optimization for
tAnt colony optimization forcsandit
 
K‐MEANS CLUSTERING ANDSNAKES PATTERN USED FOR ROAD EXTRACTION
K‐MEANS CLUSTERING ANDSNAKES PATTERN USED FOR ROAD EXTRACTIONK‐MEANS CLUSTERING ANDSNAKES PATTERN USED FOR ROAD EXTRACTION
K‐MEANS CLUSTERING ANDSNAKES PATTERN USED FOR ROAD EXTRACTIONijiert bestjournal
 
International Journal of Computational Engineering Research(IJCER)
International Journal of Computational Engineering Research(IJCER)International Journal of Computational Engineering Research(IJCER)
International Journal of Computational Engineering Research(IJCER)ijceronline
 
A New Method to Solving Generalized Fuzzy Transportation Problem-Harmonic Mea...
A New Method to Solving Generalized Fuzzy Transportation Problem-Harmonic Mea...A New Method to Solving Generalized Fuzzy Transportation Problem-Harmonic Mea...
A New Method to Solving Generalized Fuzzy Transportation Problem-Harmonic Mea...AI Publications
 
A New Method for Solving Transportation Problems Considering Average Penalty
A New Method for Solving Transportation Problems Considering Average PenaltyA New Method for Solving Transportation Problems Considering Average Penalty
A New Method for Solving Transportation Problems Considering Average PenaltyIOSRJM
 
Road surface classification based on LBP and GLCM features using kNN classifier
Road surface classification based on LBP and GLCM features using kNN classifierRoad surface classification based on LBP and GLCM features using kNN classifier
Road surface classification based on LBP and GLCM features using kNN classifierjournalBEEI
 
Emergency response behaviour data collection issue
Emergency response behaviour data collection issueEmergency response behaviour data collection issue
Emergency response behaviour data collection issueSerge Hoogendoorn
 
Forecasting electricity usage in industrial applications with gpu acceleratio...
Forecasting electricity usage in industrial applications with gpu acceleratio...Forecasting electricity usage in industrial applications with gpu acceleratio...
Forecasting electricity usage in industrial applications with gpu acceleratio...Conference Papers
 
Improving transport in Malta using GIS and LBS
Improving transport in Malta using GIS and LBSImproving transport in Malta using GIS and LBS
Improving transport in Malta using GIS and LBSMatthew Pulis
 
Camera Analytics System (Based on IEEE topic Camera Selection for adaptive hu...
Camera Analytics System (Based on IEEE topic Camera Selection for adaptive hu...Camera Analytics System (Based on IEEE topic Camera Selection for adaptive hu...
Camera Analytics System (Based on IEEE topic Camera Selection for adaptive hu...xavier_sa
 
Prediction of traveller information and route choice
Prediction of traveller information and route choicePrediction of traveller information and route choice
Prediction of traveller information and route choiceayishairshad
 
Unraveling urban traffic flows
Unraveling urban traffic flowsUnraveling urban traffic flows
Unraveling urban traffic flowsSerge Hoogendoorn
 

What's hot (16)

IRJET- Road Recognition from Remote Sensing Imagery using Machine Learning
IRJET- Road Recognition from Remote Sensing Imagery using Machine LearningIRJET- Road Recognition from Remote Sensing Imagery using Machine Learning
IRJET- Road Recognition from Remote Sensing Imagery using Machine Learning
 
paper
paperpaper
paper
 
tAnt colony optimization for
tAnt colony optimization fortAnt colony optimization for
tAnt colony optimization for
 
K‐MEANS CLUSTERING ANDSNAKES PATTERN USED FOR ROAD EXTRACTION
K‐MEANS CLUSTERING ANDSNAKES PATTERN USED FOR ROAD EXTRACTIONK‐MEANS CLUSTERING ANDSNAKES PATTERN USED FOR ROAD EXTRACTION
K‐MEANS CLUSTERING ANDSNAKES PATTERN USED FOR ROAD EXTRACTION
 
International Journal of Computational Engineering Research(IJCER)
International Journal of Computational Engineering Research(IJCER)International Journal of Computational Engineering Research(IJCER)
International Journal of Computational Engineering Research(IJCER)
 
A New Method to Solving Generalized Fuzzy Transportation Problem-Harmonic Mea...
A New Method to Solving Generalized Fuzzy Transportation Problem-Harmonic Mea...A New Method to Solving Generalized Fuzzy Transportation Problem-Harmonic Mea...
A New Method to Solving Generalized Fuzzy Transportation Problem-Harmonic Mea...
 
A real-time system for vehicle detection with shadow removal and vehicle clas...
A real-time system for vehicle detection with shadow removal and vehicle clas...A real-time system for vehicle detection with shadow removal and vehicle clas...
A real-time system for vehicle detection with shadow removal and vehicle clas...
 
A New Method for Solving Transportation Problems Considering Average Penalty
A New Method for Solving Transportation Problems Considering Average PenaltyA New Method for Solving Transportation Problems Considering Average Penalty
A New Method for Solving Transportation Problems Considering Average Penalty
 
Road surface classification based on LBP and GLCM features using kNN classifier
Road surface classification based on LBP and GLCM features using kNN classifierRoad surface classification based on LBP and GLCM features using kNN classifier
Road surface classification based on LBP and GLCM features using kNN classifier
 
Emergency response behaviour data collection issue
Emergency response behaviour data collection issueEmergency response behaviour data collection issue
Emergency response behaviour data collection issue
 
Forecasting electricity usage in industrial applications with gpu acceleratio...
Forecasting electricity usage in industrial applications with gpu acceleratio...Forecasting electricity usage in industrial applications with gpu acceleratio...
Forecasting electricity usage in industrial applications with gpu acceleratio...
 
Improving transport in Malta using GIS and LBS
Improving transport in Malta using GIS and LBSImproving transport in Malta using GIS and LBS
Improving transport in Malta using GIS and LBS
 
Camera Analytics System (Based on IEEE topic Camera Selection for adaptive hu...
Camera Analytics System (Based on IEEE topic Camera Selection for adaptive hu...Camera Analytics System (Based on IEEE topic Camera Selection for adaptive hu...
Camera Analytics System (Based on IEEE topic Camera Selection for adaptive hu...
 
Prediction of traveller information and route choice
Prediction of traveller information and route choicePrediction of traveller information and route choice
Prediction of traveller information and route choice
 
Unraveling urban traffic flows
Unraveling urban traffic flowsUnraveling urban traffic flows
Unraveling urban traffic flows
 
Rpc
RpcRpc
Rpc
 

Similar to Comp prese (1)

An IoT based Dynamic Traffic Signal Control
An IoT based Dynamic Traffic Signal ControlAn IoT based Dynamic Traffic Signal Control
An IoT based Dynamic Traffic Signal ControlGauthamSK4
 
Help the Genetic Algorithm to Minimize the Urban Traffic on Intersections
Help the Genetic Algorithm to Minimize the Urban Traffic on IntersectionsHelp the Genetic Algorithm to Minimize the Urban Traffic on Intersections
Help the Genetic Algorithm to Minimize the Urban Traffic on IntersectionsIJORCS
 
Design of intelligent traffic light controller using gsm & embedded system
Design of intelligent traffic light controller using gsm & embedded systemDesign of intelligent traffic light controller using gsm & embedded system
Design of intelligent traffic light controller using gsm & embedded systemYakkali Kiran
 
A Computational Study Of Traffic Assignment Algorithms
A Computational Study Of Traffic Assignment AlgorithmsA Computational Study Of Traffic Assignment Algorithms
A Computational Study Of Traffic Assignment AlgorithmsNicole Adams
 
A Computational Study Of Traffic Assignment Algorithms
A Computational Study Of Traffic Assignment AlgorithmsA Computational Study Of Traffic Assignment Algorithms
A Computational Study Of Traffic Assignment AlgorithmsAlicia Buske
 
Spectral opportunity selection based on the hybrid algorithm AHP-ELECTRE
Spectral opportunity selection based on the hybrid algorithm AHP-ELECTRESpectral opportunity selection based on the hybrid algorithm AHP-ELECTRE
Spectral opportunity selection based on the hybrid algorithm AHP-ELECTRETELKOMNIKA JOURNAL
 
710201909
710201909710201909
710201909IJRAT
 
A Robust Algorithm To Solve The Signal Setting Problem Considering Different ...
A Robust Algorithm To Solve The Signal Setting Problem Considering Different ...A Robust Algorithm To Solve The Signal Setting Problem Considering Different ...
A Robust Algorithm To Solve The Signal Setting Problem Considering Different ...Joshua Gorinson
 
Design and-analysis-of-a-two-stage-traffic-light-system-using-fuzzy-logic-216...
Design and-analysis-of-a-two-stage-traffic-light-system-using-fuzzy-logic-216...Design and-analysis-of-a-two-stage-traffic-light-system-using-fuzzy-logic-216...
Design and-analysis-of-a-two-stage-traffic-light-system-using-fuzzy-logic-216...hanhdoduc
 
Urban Bus Route Planning Using Reverse Labeling Dijkstra Algorithm for Tempor...
Urban Bus Route Planning Using Reverse Labeling Dijkstra Algorithm for Tempor...Urban Bus Route Planning Using Reverse Labeling Dijkstra Algorithm for Tempor...
Urban Bus Route Planning Using Reverse Labeling Dijkstra Algorithm for Tempor...IRJET Journal
 
A Review on Traffic Signal Identification
A Review on Traffic Signal IdentificationA Review on Traffic Signal Identification
A Review on Traffic Signal Identificationijtsrd
 
Application of-computational-intelligence-techniques-for-economic-load-dispatch
Application of-computational-intelligence-techniques-for-economic-load-dispatchApplication of-computational-intelligence-techniques-for-economic-load-dispatch
Application of-computational-intelligence-techniques-for-economic-load-dispatchCemal Ardil
 
Simulation analysis Halmstad University 2013_project
Simulation analysis Halmstad University 2013_projectSimulation analysis Halmstad University 2013_project
Simulation analysis Halmstad University 2013_projectAlexandru Gutu
 
A Hybrid Method for Automatic Traffic Control Mechanism
A Hybrid Method for Automatic Traffic Control MechanismA Hybrid Method for Automatic Traffic Control Mechanism
A Hybrid Method for Automatic Traffic Control MechanismMangaiK4
 
Conflict-free dynamic route multi-agv using dijkstra Floyd-warshall hybrid a...
Conflict-free dynamic route multi-agv using dijkstra  Floyd-warshall hybrid a...Conflict-free dynamic route multi-agv using dijkstra  Floyd-warshall hybrid a...
Conflict-free dynamic route multi-agv using dijkstra Floyd-warshall hybrid a...IJECEIAES
 
Vehicle route scheduling and transportation cost minimization in a latex indu...
Vehicle route scheduling and transportation cost minimization in a latex indu...Vehicle route scheduling and transportation cost minimization in a latex indu...
Vehicle route scheduling and transportation cost minimization in a latex indu...IJRES Journal
 
Optimization of Corridor Observation Method to Solve Environmental and Econom...
Optimization of Corridor Observation Method to Solve Environmental and Econom...Optimization of Corridor Observation Method to Solve Environmental and Econom...
Optimization of Corridor Observation Method to Solve Environmental and Econom...ijceronline
 
A two Stage Fuzzy Logic Adaptive Traffic Signal Control for an Isolated Inter...
A two Stage Fuzzy Logic Adaptive Traffic Signal Control for an Isolated Inter...A two Stage Fuzzy Logic Adaptive Traffic Signal Control for an Isolated Inter...
A two Stage Fuzzy Logic Adaptive Traffic Signal Control for an Isolated Inter...ijtsrd
 

Similar to Comp prese (1) (20)

An IoT based Dynamic Traffic Signal Control
An IoT based Dynamic Traffic Signal ControlAn IoT based Dynamic Traffic Signal Control
An IoT based Dynamic Traffic Signal Control
 
Help the Genetic Algorithm to Minimize the Urban Traffic on Intersections
Help the Genetic Algorithm to Minimize the Urban Traffic on IntersectionsHelp the Genetic Algorithm to Minimize the Urban Traffic on Intersections
Help the Genetic Algorithm to Minimize the Urban Traffic on Intersections
 
Design of intelligent traffic light controller using gsm & embedded system
Design of intelligent traffic light controller using gsm & embedded systemDesign of intelligent traffic light controller using gsm & embedded system
Design of intelligent traffic light controller using gsm & embedded system
 
A Computational Study Of Traffic Assignment Algorithms
A Computational Study Of Traffic Assignment AlgorithmsA Computational Study Of Traffic Assignment Algorithms
A Computational Study Of Traffic Assignment Algorithms
 
A Computational Study Of Traffic Assignment Algorithms
A Computational Study Of Traffic Assignment AlgorithmsA Computational Study Of Traffic Assignment Algorithms
A Computational Study Of Traffic Assignment Algorithms
 
Spectral opportunity selection based on the hybrid algorithm AHP-ELECTRE
Spectral opportunity selection based on the hybrid algorithm AHP-ELECTRESpectral opportunity selection based on the hybrid algorithm AHP-ELECTRE
Spectral opportunity selection based on the hybrid algorithm AHP-ELECTRE
 
710201909
710201909710201909
710201909
 
A Robust Algorithm To Solve The Signal Setting Problem Considering Different ...
A Robust Algorithm To Solve The Signal Setting Problem Considering Different ...A Robust Algorithm To Solve The Signal Setting Problem Considering Different ...
A Robust Algorithm To Solve The Signal Setting Problem Considering Different ...
 
Design and-analysis-of-a-two-stage-traffic-light-system-using-fuzzy-logic-216...
Design and-analysis-of-a-two-stage-traffic-light-system-using-fuzzy-logic-216...Design and-analysis-of-a-two-stage-traffic-light-system-using-fuzzy-logic-216...
Design and-analysis-of-a-two-stage-traffic-light-system-using-fuzzy-logic-216...
 
Urban Bus Route Planning Using Reverse Labeling Dijkstra Algorithm for Tempor...
Urban Bus Route Planning Using Reverse Labeling Dijkstra Algorithm for Tempor...Urban Bus Route Planning Using Reverse Labeling Dijkstra Algorithm for Tempor...
Urban Bus Route Planning Using Reverse Labeling Dijkstra Algorithm for Tempor...
 
A Review on Traffic Signal Identification
A Review on Traffic Signal IdentificationA Review on Traffic Signal Identification
A Review on Traffic Signal Identification
 
Application of-computational-intelligence-techniques-for-economic-load-dispatch
Application of-computational-intelligence-techniques-for-economic-load-dispatchApplication of-computational-intelligence-techniques-for-economic-load-dispatch
Application of-computational-intelligence-techniques-for-economic-load-dispatch
 
Simulation analysis Halmstad University 2013_project
Simulation analysis Halmstad University 2013_projectSimulation analysis Halmstad University 2013_project
Simulation analysis Halmstad University 2013_project
 
C41016023
C41016023C41016023
C41016023
 
A Hybrid Method for Automatic Traffic Control Mechanism
A Hybrid Method for Automatic Traffic Control MechanismA Hybrid Method for Automatic Traffic Control Mechanism
A Hybrid Method for Automatic Traffic Control Mechanism
 
Conflict-free dynamic route multi-agv using dijkstra Floyd-warshall hybrid a...
Conflict-free dynamic route multi-agv using dijkstra  Floyd-warshall hybrid a...Conflict-free dynamic route multi-agv using dijkstra  Floyd-warshall hybrid a...
Conflict-free dynamic route multi-agv using dijkstra Floyd-warshall hybrid a...
 
Timetable synchronization
Timetable synchronization Timetable synchronization
Timetable synchronization
 
Vehicle route scheduling and transportation cost minimization in a latex indu...
Vehicle route scheduling and transportation cost minimization in a latex indu...Vehicle route scheduling and transportation cost minimization in a latex indu...
Vehicle route scheduling and transportation cost minimization in a latex indu...
 
Optimization of Corridor Observation Method to Solve Environmental and Econom...
Optimization of Corridor Observation Method to Solve Environmental and Econom...Optimization of Corridor Observation Method to Solve Environmental and Econom...
Optimization of Corridor Observation Method to Solve Environmental and Econom...
 
A two Stage Fuzzy Logic Adaptive Traffic Signal Control for an Isolated Inter...
A two Stage Fuzzy Logic Adaptive Traffic Signal Control for an Isolated Inter...A two Stage Fuzzy Logic Adaptive Traffic Signal Control for an Isolated Inter...
A two Stage Fuzzy Logic Adaptive Traffic Signal Control for an Isolated Inter...
 

Recently uploaded

DevoxxFR 2024 Reproducible Builds with Apache Maven
DevoxxFR 2024 Reproducible Builds with Apache MavenDevoxxFR 2024 Reproducible Builds with Apache Maven
DevoxxFR 2024 Reproducible Builds with Apache MavenHervé Boutemy
 
Transcript: New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024
Transcript: New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024Transcript: New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024
Transcript: New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024BookNet Canada
 
SIP trunking in Janus @ Kamailio World 2024
SIP trunking in Janus @ Kamailio World 2024SIP trunking in Janus @ Kamailio World 2024
SIP trunking in Janus @ Kamailio World 2024Lorenzo Miniero
 
"Debugging python applications inside k8s environment", Andrii Soldatenko
"Debugging python applications inside k8s environment", Andrii Soldatenko"Debugging python applications inside k8s environment", Andrii Soldatenko
"Debugging python applications inside k8s environment", Andrii SoldatenkoFwdays
 
Unraveling Multimodality with Large Language Models.pdf
Unraveling Multimodality with Large Language Models.pdfUnraveling Multimodality with Large Language Models.pdf
Unraveling Multimodality with Large Language Models.pdfAlex Barbosa Coqueiro
 
Install Stable Diffusion in windows machine
Install Stable Diffusion in windows machineInstall Stable Diffusion in windows machine
Install Stable Diffusion in windows machinePadma Pradeep
 
The Future of Software Development - Devin AI Innovative Approach.pdf
The Future of Software Development - Devin AI Innovative Approach.pdfThe Future of Software Development - Devin AI Innovative Approach.pdf
The Future of Software Development - Devin AI Innovative Approach.pdfSeasiaInfotech2
 
Streamlining Python Development: A Guide to a Modern Project Setup
Streamlining Python Development: A Guide to a Modern Project SetupStreamlining Python Development: A Guide to a Modern Project Setup
Streamlining Python Development: A Guide to a Modern Project SetupFlorian Wilhelm
 
My INSURER PTE LTD - Insurtech Innovation Award 2024
My INSURER PTE LTD - Insurtech Innovation Award 2024My INSURER PTE LTD - Insurtech Innovation Award 2024
My INSURER PTE LTD - Insurtech Innovation Award 2024The Digital Insurer
 
Vector Databases 101 - An introduction to the world of Vector Databases
Vector Databases 101 - An introduction to the world of Vector DatabasesVector Databases 101 - An introduction to the world of Vector Databases
Vector Databases 101 - An introduction to the world of Vector DatabasesZilliz
 
DevEX - reference for building teams, processes, and platforms
DevEX - reference for building teams, processes, and platformsDevEX - reference for building teams, processes, and platforms
DevEX - reference for building teams, processes, and platformsSergiu Bodiu
 
My Hashitalk Indonesia April 2024 Presentation
My Hashitalk Indonesia April 2024 PresentationMy Hashitalk Indonesia April 2024 Presentation
My Hashitalk Indonesia April 2024 PresentationRidwan Fadjar
 
Human Factors of XR: Using Human Factors to Design XR Systems
Human Factors of XR: Using Human Factors to Design XR SystemsHuman Factors of XR: Using Human Factors to Design XR Systems
Human Factors of XR: Using Human Factors to Design XR SystemsMark Billinghurst
 
Ensuring Technical Readiness For Copilot in Microsoft 365
Ensuring Technical Readiness For Copilot in Microsoft 365Ensuring Technical Readiness For Copilot in Microsoft 365
Ensuring Technical Readiness For Copilot in Microsoft 3652toLead Limited
 
Leverage Zilliz Serverless - Up to 50X Saving for Your Vector Storage Cost
Leverage Zilliz Serverless - Up to 50X Saving for Your Vector Storage CostLeverage Zilliz Serverless - Up to 50X Saving for Your Vector Storage Cost
Leverage Zilliz Serverless - Up to 50X Saving for Your Vector Storage CostZilliz
 
CloudStudio User manual (basic edition):
CloudStudio User manual (basic edition):CloudStudio User manual (basic edition):
CloudStudio User manual (basic edition):comworks
 
Training state-of-the-art general text embedding
Training state-of-the-art general text embeddingTraining state-of-the-art general text embedding
Training state-of-the-art general text embeddingZilliz
 
New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024
New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024
New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024BookNet Canada
 

Recently uploaded (20)

DevoxxFR 2024 Reproducible Builds with Apache Maven
DevoxxFR 2024 Reproducible Builds with Apache MavenDevoxxFR 2024 Reproducible Builds with Apache Maven
DevoxxFR 2024 Reproducible Builds with Apache Maven
 
Transcript: New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024
Transcript: New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024Transcript: New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024
Transcript: New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024
 
SIP trunking in Janus @ Kamailio World 2024
SIP trunking in Janus @ Kamailio World 2024SIP trunking in Janus @ Kamailio World 2024
SIP trunking in Janus @ Kamailio World 2024
 
"Debugging python applications inside k8s environment", Andrii Soldatenko
"Debugging python applications inside k8s environment", Andrii Soldatenko"Debugging python applications inside k8s environment", Andrii Soldatenko
"Debugging python applications inside k8s environment", Andrii Soldatenko
 
Unraveling Multimodality with Large Language Models.pdf
Unraveling Multimodality with Large Language Models.pdfUnraveling Multimodality with Large Language Models.pdf
Unraveling Multimodality with Large Language Models.pdf
 
Install Stable Diffusion in windows machine
Install Stable Diffusion in windows machineInstall Stable Diffusion in windows machine
Install Stable Diffusion in windows machine
 
The Future of Software Development - Devin AI Innovative Approach.pdf
The Future of Software Development - Devin AI Innovative Approach.pdfThe Future of Software Development - Devin AI Innovative Approach.pdf
The Future of Software Development - Devin AI Innovative Approach.pdf
 
Streamlining Python Development: A Guide to a Modern Project Setup
Streamlining Python Development: A Guide to a Modern Project SetupStreamlining Python Development: A Guide to a Modern Project Setup
Streamlining Python Development: A Guide to a Modern Project Setup
 
My INSURER PTE LTD - Insurtech Innovation Award 2024
My INSURER PTE LTD - Insurtech Innovation Award 2024My INSURER PTE LTD - Insurtech Innovation Award 2024
My INSURER PTE LTD - Insurtech Innovation Award 2024
 
Vector Databases 101 - An introduction to the world of Vector Databases
Vector Databases 101 - An introduction to the world of Vector DatabasesVector Databases 101 - An introduction to the world of Vector Databases
Vector Databases 101 - An introduction to the world of Vector Databases
 
DevEX - reference for building teams, processes, and platforms
DevEX - reference for building teams, processes, and platformsDevEX - reference for building teams, processes, and platforms
DevEX - reference for building teams, processes, and platforms
 
My Hashitalk Indonesia April 2024 Presentation
My Hashitalk Indonesia April 2024 PresentationMy Hashitalk Indonesia April 2024 Presentation
My Hashitalk Indonesia April 2024 Presentation
 
Human Factors of XR: Using Human Factors to Design XR Systems
Human Factors of XR: Using Human Factors to Design XR SystemsHuman Factors of XR: Using Human Factors to Design XR Systems
Human Factors of XR: Using Human Factors to Design XR Systems
 
Ensuring Technical Readiness For Copilot in Microsoft 365
Ensuring Technical Readiness For Copilot in Microsoft 365Ensuring Technical Readiness For Copilot in Microsoft 365
Ensuring Technical Readiness For Copilot in Microsoft 365
 
Leverage Zilliz Serverless - Up to 50X Saving for Your Vector Storage Cost
Leverage Zilliz Serverless - Up to 50X Saving for Your Vector Storage CostLeverage Zilliz Serverless - Up to 50X Saving for Your Vector Storage Cost
Leverage Zilliz Serverless - Up to 50X Saving for Your Vector Storage Cost
 
CloudStudio User manual (basic edition):
CloudStudio User manual (basic edition):CloudStudio User manual (basic edition):
CloudStudio User manual (basic edition):
 
Training state-of-the-art general text embedding
Training state-of-the-art general text embeddingTraining state-of-the-art general text embedding
Training state-of-the-art general text embedding
 
DMCC Future of Trade Web3 - Special Edition
DMCC Future of Trade Web3 - Special EditionDMCC Future of Trade Web3 - Special Edition
DMCC Future of Trade Web3 - Special Edition
 
E-Vehicle_Hacking_by_Parul Sharma_null_owasp.pptx
E-Vehicle_Hacking_by_Parul Sharma_null_owasp.pptxE-Vehicle_Hacking_by_Parul Sharma_null_owasp.pptx
E-Vehicle_Hacking_by_Parul Sharma_null_owasp.pptx
 
New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024
New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024
New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024
 

Comp prese (1)

  • 1. Intelligent Traffic Light Control Mohamed Khasawneh CIISE, Concordia University PhD Comprehensive Exam 1
  • 2. Outline 1. Introduction 2. Context 3. Papers 1. Intelligent Traffic Light Control Using Distributed Multi-agent Q Learning 2. Intelligent Traffic Lights: Green Time Period Negotiation 3. Optimal Cycle Program of Traffic Lights with Particle Swarm Optimization 4. Conclusions 5. Timeline 6. Future works 2
  • 3. 1. Introduction • Nowadays, studying Intelligent Transport Systems (ITS) is considered to be one of the hottest topics which can improve the quality of life in many ways. Designing a good ITS system can bring us many benefits such as saving lives by reducing the percentage of accidents, increasing the safety for pedestrians, reducing the waiting times for both vehicles and pedestrians, optimizing fuel consumption, reducing the emissions of CO and CO2 gases, reducing the noise pollution, etc. 3
  • 4. 2. Context • Smart cities • Traffic Congestion management problem • Intelligent Traffic signaling problem • Connected vehicles 4
  • 5. Challenges • There are many challenges in designing these systems: The lack of global view of traffic conditions at the neighboring streets or intersections. The unexpected change in traffic condition at any given time. The high congestion status which makes it a hard and daunting task to accomplish optimization between vehicles and pedestrian in a fair and satisfying manner. 5
  • 6. Papers under study 1. Intelligent Traffic Light Control Using Distributed Multi-agent Q Learning. 2. Intelligent Traffic Lights: Green Time Period Negotiation. 3. Optimal Cycle Program of Traffic Lights with Particle Swarm Optimization. 6
  • 7. Paper 1. Intelligent Traffic Light Control Using Distributed Multi-agent Q Learning 7
  • 8. Problem Definition • Designing Intelligent traffic light control, where two main aspects are not considered in the previous studies: Non motorized traffic. Global optimization. 8
  • 9. Literature Review 9 Figure 1.Classification Of Traffic Light Control Approach
  • 10. Proposed System Design • The proposed system design is based on the usage of Q learning which falls under the reinforcement learning techniques. • Q learning technique is composed of two main phases which are exploration and exploitation phases. • In exploration phase, the Q learning agent is randomly exploring the environment around, after the environment is learned an exploitation phase is starting with an objective to maximize the rewards in the whole system. 10
  • 11. Proposed System Design Figure 2. System Architecture for Distributed Q Learning Agent [2]. 11
  • 12. Methodology • A Markov Decision Based model where states, actions, rewards are defined. States : 𝑆𝑖,𝑑 𝑡 = {𝑞1𝑖,𝑑 𝑡 , 𝑞2𝑖,𝑑 𝑡 , … , 𝑞 𝑗𝑖,𝑑 𝑡 , 𝑚1𝑖,𝑑,𝐿 𝑡 , 𝑚1𝑖,𝑑, 𝑡 , … , 𝑚1𝑖,𝑑 𝑡 , 𝑚1𝑖,𝑑 𝑡 } (1) Actions: Fig. 3: Action sets (green lights) for “+” shape intersection 12
  • 13. Methodology Rewards: 𝑹𝒊,𝒅 𝒕 𝒂𝒊,𝒅 𝒕 , 𝒂𝒋,𝒅 𝒕 , 𝑺𝒊,𝒅 𝒕 , 𝑺𝒋,𝒅, 𝒕 𝑾𝒊,𝒅 𝒕 = −( 𝑾 𝟏,𝒅 𝒕 𝑵 𝒊 𝒋𝜺𝑵 𝒊 𝒒𝒋𝒊,𝒅 𝒕 + 𝑾 𝒛,𝒅 𝒕 𝑵 𝒊 𝑵 𝒋 𝒋𝜺𝑵 𝒊 𝒌𝜺𝑵 𝒋 𝒒 𝒌𝒋,𝒅 𝒕 + 𝑾 𝟑,𝒅 𝒕 𝟐 𝑵 𝒊 𝒋𝜺𝑵 𝒊 ( 𝒎𝒋𝒊,𝒅,𝑳 𝒕 + 𝒎𝒋𝒊,𝒅,𝑹 𝒕 )) (2) Integrate real-world constraints and traffic rules. 13
  • 14. Used parameters Parameters Meaning 𝑺𝒊,𝒅 𝒕 The state at intersection i, at day d and time t. 𝒒 𝟏𝒊,𝒅 𝒕 , 𝒒 𝟐𝒊,𝒅 𝒕 , … , 𝒒𝒋𝒊,𝒅 𝒕 The queue length from intersection j to intersection i, at day d and time t 𝒎 𝟏𝒊,𝒅,𝑳 𝒕 , 𝒎 𝟏𝒊,𝒅,𝑹 𝒕 The queue length for pedestrians at the left side from intersection j to i, at day d and time t, 𝑹𝒊,𝒅 𝒕 The reward at intersection i, at day d and time 𝑾 𝟏,𝒅 𝒕 The weight to present the local vehicular queues at intersection I 𝑾 𝟐,𝒅 𝒕 The weight to present the neighborhood vehicular queues at the neighbors of intersection I 𝑾 𝟑,𝒅 𝒕 Weight to present the total pedestrian queues at intersection I 𝒂𝒊,𝒅 𝒕 The action at intersection i, at day d and time t 14Table 1. Used Parameters
  • 15. Results Fig. 4: Total cumulated queue lengths for vehicles and pedestrians in the case of high pedestrian rate Fig. 5: Total waiting and traveling time for vehicles and pedestrians in the case of high pedestrian rate 15
  • 16. Results Fig. 6: CO2 and CO emission of vehicles in the case of high pedestrian rate Fig. 7: Fuel consumption and noise pollution of vehicles in the case of high pedestrian rate Fig. 8: Total cumulated queue lengths for vehicles and pedestrians in the case of medium pedestrian rate Fig. 9: Total waiting and traveling time for vehicles and pedestrians in the case of medium pedestrian rate 16
  • 17. Conclusion • Unlike other papers in the literature, one of the main contributions of this paper is handling both motorized and non-motorized vehicles. Non-motorized traffic such as pedestrian and bicyclist are also an important part of the whole system and should not be overlooked in order to achieve the fairness among all users. Other contribution is the achievement of near global optimal solution by gathering/exchanging all the information from/to local and neighboring intersections, this helps in a better and more accurate decisions compared with other proposed algorithms in literature which are either using fixed-time or dynamic control approach 17
  • 18. Paper 2: Intelligent Traffic Lights: Green Time Period Negotiation 18
  • 19. Problem Definition • Designing Intelligent traffic light control, where two main aspects are not considered in the previous studies: Person-based. The distribution of green time period between the traffic streams. 19
  • 20. Literature Review 20 Author Approach Dresner and Stone (2005) (2009) “reservation based” strategy Vasirani and Ossowski(2009) (2011) “market based” strategy and extended the work done by Dresner and Stone Schepperle and Bohm(2008) this strategy will take in account valuation of the driver Table 2. Literature Review
  • 21. Proposed System Design • A multi-agent computerized system (MAS or "self-organized system") composed of multiple interacting intelligent agents is proposed. Two stages are involved: First stage : try to find a traffic plan through defining the phase and determining green time periods Second stage: include two decisions defining the time of terminating the current stage and defining the next stage to be implemented. 21
  • 22. First stage 𝑵 𝟎 = 𝑸𝑻 𝒇 𝟒 (𝑿𝒊−𝟏) + 𝒙𝒊 − 𝟏 𝟐 + 𝟏𝟐 𝒙 𝒊− .𝟔𝟕+ 𝒔𝒈 𝟔𝟎𝟎 𝑸𝑻 𝒇 𝒊𝒇 𝑿𝒊 >. 𝟔𝟕 + 𝒔𝒈 𝟔𝟎𝟎 (3) 0 otherwise 𝑵 𝑴𝒂𝒙 = 𝒒 𝑪−𝒈 𝟏−𝒚 + 𝑵 𝟎 𝒏𝒖𝒎𝒃𝒆𝒓 𝒐𝒇 𝒍𝒂𝒏𝒆𝒔 < 𝒍𝒂𝒏𝒆 𝒄𝒂𝒑𝒂𝒄𝒊𝒕𝒚 𝒒𝒖𝒆𝒖𝒆 (4) 𝑫 𝒕𝒐𝒕𝒂𝒍=𝒗𝒊𝒆𝒄𝒉𝒍𝒆𝒔+𝒑𝒆𝒅𝒆𝒔𝒕𝒓𝒂𝒊𝒏𝒔 = 𝒊=𝟏 𝑰 𝒒𝑪(𝟏−𝒈𝒊/𝑪) 𝟐 𝟐(𝟏−𝒙 𝒊 (𝒈𝒊/𝑪) + 𝑵 𝟎 𝑿𝒊 𝒗𝒆𝒉 𝒐𝒄𝒄 (5) 22
  • 25. Case study Variables values Simulation Time 2 hours The inter-green time 5s (3s yellow, 2s red). The minimum green duration 7 second The maximum planned cycle length 120 second the maximum acceptable degree of saturation 80% The saturation flows for straight-ahead movements, right-turn movement and left-turn movements 1800, 1600 and 1700 veh/h/lane, respectively. average vehicle occupancy 1.2 Simulation Software AIMSUN Table 3. Simulation Parameters 25
  • 26. Result Fig. 5 – Average travel time of each replication versus time: a) baseline; b) proposed 26
  • 27. Conclusions • This paper unveiled a person-oriented traffic signal control method for secluded intersection that include signal timing optimization and concurrent signal plan designs with actuated-time data about the dynamics of the network. • The suggested control mechanism operates within two phases: first, it chooses traffic signals and second, it undertakes negotiation of green time duration based on actual traffic situations. The suggested control is assessed using a tiny traffic simulation tool. 27
  • 28. Paper 3: Optimal Cycle Program of Traffic Lights With Particle Swarm Optimization 28
  • 29. Problem definition • Designing Intelligent traffic light control, where three main aspects are not considered in the previous studies: Large vehicular networks. Many scenarios. Comparison with other algorithms. 29
  • 30. Literature Review Mathematic models Fuzzy logic approaches Biologically inspired optimizers McCrea and Moutari[5]( merged knowledge-founded and continuous calculus-founded models with the intent of defining the road networks’ traffic flow)  Tolba et al.( offered a Petri net- based model to indicate traffic flow, from both macroscopic and macroscopic perceptions, consideration of vehicles’ individual trajectories and observation of global variables, respectively Lim et al. [4]( derived a fuzzy logic controller during a single intersection’s real-time local optimization)  Other experts [3] developed a traffic simulator utilizing fuzzy logic agents for traffic lights at isolated intersections. Cellular automata:  Brockfeld et al. [6] utilized a CA model whereby the city network was instigated as a simple square having some normal streets and four junctions. Neural networks (NN):  Spall and Chin [1] illustrated an NN considered in configuring traffic lights’ control parameters. Genetics Algorithm 30 Table 4.Literature Review(Paper 2)
  • 31. Solution Approach • PSO has been considered in [7] and [8] as a population-based meta-heuristic motivated by the birds’ social behavior within a flock, whose initial purpose was to continually optimize problems. • In this model, initial solutions are randomly generated .each solutions to the problem is identified as particle locations while the particles’ population is known as swarm. For this algorithm, xi, the position of each particle, is updated for each iteration g as follows: where term 𝑣 𝑔+1 𝑖 is the velocity of the particle, given by: 31 𝑋 𝑔+1 𝑖 = 𝑋 𝑔 𝑖 + 𝑣 𝑔+1 𝑖 𝑣 𝑔+1 𝑖 = 𝑤. 𝑣𝑔 𝑖 +U[0,𝜑1]. (𝑃𝑔 𝑖 − 𝑥 𝑔 𝑖 )+U[0,𝜑2]. 𝑏 𝑔 − 𝑥 𝑔 𝑖 .
  • 33. PSO for Traffic Light Scheduling The optimization solver proposed in this paper for optimal cycle programs for traffic light involves three steps which are : Solution Encoding: all traffic light logics with their phase durations are mapped into each solution and those solutions are randomly generated by PSO The fitness function :Each solution is evaluated, taking into account the information obtained from events that occur during simulation by the following equation: F(s)= 𝑣=0 𝑉 𝑗 𝑣 𝑠 + 𝑣=0 𝑉+𝐶 𝑤 𝑣 𝑠 +(𝐶 𝑠 .𝑆𝑡) 𝑉2 𝑠 +𝐶𝑟 The global optimization procedure: the optimization strategy is composed into two parts : an optimization algorithm a simulation procedure 33
  • 34. Case study 34 Table 5. Simulation and PSO Parameters
  • 36. Papers comparison Weakness Strengths Paper 1 o Vehicle-based strategy o does not react well to a sharp changes in traffic patterns o Global optimization o Motorized and non-motorized traffic Paper 2 o The proposed mechanism is only admissible where the demand of pedestrians is low. o Person-based strategy o the system is flexible Paper 3 o priority for motorized traffic. o unlimited vehicular networks o various scenario topologies o Compared with four algorithms 36 Table 6. Papers comparison
  • 37. Conclusions • In this report, research papers tackling the efficient design of intelligent traffic light are discussed. Considering different parameters to reach a different objective function. • In [2], both motorized and non-motorized vehicles are taken into account to minimize the overall congestion and waiting time. Whereas, in [9] an auction based intersection control strategy is proposed with the objective function of minimizing the waiting time for each person by taking into consideration the number of persons in each car. • On the other hand, in another paper used an optimization method called Particle Swarm Optimization (PSO) to find successful traffic light cycle programs with two main objectives which are the number of vehicles arriving at the destination and the total travel time. 37
  • 39. Future works • Designing a person based globally optimal intelligent traffic light which fairly considers many parameters (motorized and non-motorized traffic). • Designing a machine learning-based algorithm to predict traffic congestion and signal length timings based on historical information 39
  • 40. References [1] Y. Feng, K. L. Head, S. Khoshmagham, and M. Zamanipour, ―A real-time adaptive signal control in a connected vehicle environment,Transportation Research Part C: Emerging Technologies, vol. 55, pp. 460–473, 2015. [2] Ying Liu1, 2, Lei Liu1, Wei-Peng Chen1,Intelligent Traffic Light Control Using Distributed Multi- Agent Q Learning, Fujitsu Laboratories of America, Inc., Sunnyvale, CA, USA [3] Timóteo, I., Araújo, M., Rossetti, R. & Oliveira, E. 2010. TraSMAPI: An API oriented towards Multi- Agent Systems real-time interaction with multiple Traffic Simulators. In: 13th Intern. IEEE Conference on Intelligent Transportation Systems (ITSC), Funchal, pp. 1183-1188. [4] Dion, F., Rakha, H. & Kang, Y.-S. 2004. Comparison of delay estimates at under-saturated and over-saturated pre-timed signalized intersections. Transportation Research Part B: Methodological, 38(2):99-122. [5] J. McCrea and S. Moutari, ―A hybrid macroscopic-based model for traffic flow in road networks,Eur. J. Oper. Res., vol. 207, no. 1, pp. 676–684, 2010. 40
  • 41. References [6] R. Bretherton, N. Hounsell, and B. Radia. (1996). Public transport priority in SCOOT [Online]. Available: http://eprints.soton.ac.uk/75299/ [7] M. A. M. de Oca, T. St¨utzle, M. Birattari, and M. Dorigo, ―Frankenstein‘s PSO: A composite particle swarm optimization algorithm,IEEE Trans. Evol. Comput., vol. 13, no. 5, pp. 1120–1132, Oct. 2009. [8] J. Kennedy and R. C. Eberhart, Swarm Intelligence. San Francisco, CA, USA: Morgan Kaufmann, 2001. [9] Cristina Vilarinho, José Pedro Tavares, Rosaldo J. F. Rossetti, Intelligent Traffic Lights: Green Time Period Negotiato. 41