Intelligent Project (Final)

7/28/2019 Intelligent Project (Final)

1/56

1

Faculty of Engineering & Built Environment

KAAA 6424

INTELLIGENT URBAN TRAFFIC CONTROL SYSTEM

PROPOSED KAJANG URBAN TRAFFIC MANAGEMENT

SYSTEM

Supervisor

Prof. Ir. Dr. Riza Atiq Bin O.K. Rahmat

Prepared by :

1-Haider Farhan P654052-Mustafa Talib P609153-SaharAbd Ali P65295


2/56

2

Table of Content

Content Page

Table of Content

List of Figure

List of Table

2

4

5

1.0 Introduction

1.1Problem Statement1.2Study Objective1.3Study Focus Area1.4Scope of Work

1.4.1 Site Visit and Visual Appraisal1.4.2 Traffic Survey and Analysis1.4.3 Output

6

6

6

7

8

8

8

8

2.0 Study Methodology

2.1 Data Collection (Traffic Survey)

2.1.1 Classified Volumetric Count

2.1.2 Travel Time Survey

2.1.3 Queue Length and Delay

2.2 Phasing Sequences Determination2.3 Determination of optimum Cycle Time and Optimum

Green Time Split

2.4 Determination of Offset

9

10

10

11

11

1112

14

3.0 Proposed Urban Traffic Management System

3.1 Proposed Traffic Control Centre

16

17


3/56


4/56


5/56


6/56

6

1.0 INTRODUCTION

Kajang, is a town in the eastern part ofSelangor, Malaysia. Kajang is the

district capital ofHulu Langat. It is located 21 kilometers (13 mi) from

Malaysia's capital, Kuala Lumpur.The current locational gravity of growth in Kajang would be Sungai

Chua. The total population of Kajang has grown rapidly in the past fewyears, with estimated population growth of 9% per annum. The soon-to-

be-realisedKlang Valley MRT station in Bandar Kajang will boost the

property value in Sungai Chua.

As of 2004, a few townships have been developed in Kajang, such asTaman Prima Saujana (straight from JalanCheras), Sungai Chua, Taman

Kajang Perdana (Kajang Highlands). Lately, many high-end

developments has mushroomed in Kajang such as Twin Palms, SriBanyan, Country Heights, Jade Hills and Prima Paramount.

Areas surrounding these new townships are easily accessible via

the SILK Expressway. Kajang is governed by the MajlisPerbandaran

Kajang.

Kajang town has grown rapidly in the past several decades. New

suburban areas and satellite townships have radiated out of the old

Kajang town. Continuous traffic growth through developed areas and

difficulties in building new transportation infrastructure have caused aneed for careful monitoring of operating conditions on existing

transportation facilities. New strategies for traffic control must be

developed in order to manage the increase in traffic volume in Kajang.

1.1 PROBLEM STATEMENT

Traffic congestion and long queues at intersections during peak hours is

the major problems in Kajang. Growing numbers of road users and

limited resources provided by current infrastructures lead to increasing of

traveling times. This problem is mainly due to poor coordination between

adjacent traffic signal controls, resulting in inefficient progressive traffic

flows (or commonly known as the unattainable green wave effect).

Inability of existing method in determining traffic demand and provide

suitable time split when the traffic volume exceeds its capacity is anothermain factor which lead to traffic congestion.

1.2 OBJECTIVES OF STUDY
http://en.wikipedia.org/wiki/Selangorhttp://en.wikipedia.org/wiki/Malaysiahttp://en.wikipedia.org/wiki/Hulu_Langathttp://en.wikipedia.org/wiki/Kuala_Lumpurhttp://en.wikipedia.org/w/index.php?title=Sungai_Chua&action=edit&redlink=1http://en.wikipedia.org/w/index.php?title=Sungai_Chua&action=edit&redlink=1http://en.wikipedia.org/wiki/Kajang_Dispersal_Link_Expresswayhttp://en.wikipedia.org/wiki/Majlis_Perbandaran_Kajanghttp://en.wikipedia.org/wiki/Majlis_Perbandaran_Kajanghttp://en.wikipedia.org/wiki/Majlis_Perbandaran_Kajanghttp://en.wikipedia.org/wiki/Majlis_Perbandaran_Kajanghttp://en.wikipedia.org/wiki/Kajang_Dispersal_Link_Expresswayhttp://en.wikipedia.org/w/index.php?title=Sungai_Chua&action=edit&redlink=1http://en.wikipedia.org/w/index.php?title=Sungai_Chua&action=edit&redlink=1http://en.wikipedia.org/wiki/Kuala_Lumpurhttp://en.wikipedia.org/wiki/Hulu_Langathttp://en.wikipedia.org/wiki/Malaysiahttp://en.wikipedia.org/wiki/Selangor


7/56

7

The main objective of the study is to propose solution to traffic

congestion in Kajang by optimizing traffic flows along a few selected

arterial routes. The solution shall include:

To upgrade current situation of traffic flow in Kajang town.To ensure the safety of the traffic users.To give real time information to the users.To propose mechanism of action during incident/breakdown.To have batter service of traffic in Kajang town.

1.3 AREA STUDY

Figure 1.1: The location of study focus area.

The study has focus on the selected location such in Figure 1.1. The study

locations are recognized as the followings:

Intersection 1:intersection linking Kajang, Babgi, and (UKM).

Intersection 2:links Kajang, UKM, and Sepakat.


8/56

8

Intersection 3:intersection which links Kajang, UKM, and the

highway

The selection of these intersections is based on the present condition of

the route. These routes recognized as roads which has huge traffic volumebut with minimum traffic control facilities. Traffic congestion occurred

everyday during peak hours in the morning, afternoon and evening. So,

this study is purposely conducted in order to evaluate the problems and to

introduce solutions to the problems.

Figure 1.2: Traffic condition atJalanreko intersection.

1.4 SCOPE OF WORK

The scope of work of the study consists of the following:

1.4.1 Site Visit and Visual Appraisal

Site visit are conducted during the study in order to get first hand

knowledge of the study focus area and to evaluate the actual site

problems. Existing traffic data, documents and drawings was examined to

obtain the information of the selected intersections. Preliminary data suchas number of lanes, distance between intersection, phasing sequences,

signal timing, and traffic volume are collected during the site visit. This

information is important for future planning.

1.4.2 Traffic Survey & Analysis

Traffic surveys are conducted at appropriate time and durations (duringpeak hour), so that actual data of traffic flows at the selected


9/56

9

intersection could be obtained. Classified Volume Count, Travel Time

Survey, and Queue Length and Delay Survey are three approaches that

been used for the traffic surveys.

Based on the traffic data and other site information collected, analysis

have been conducted to determine the best control methodology for the

intersection and to obtain the suitable optimum cycle time, green time

split and also the offset time for the study focus area.

1.4.3 Output

The output of the study consists of the followings:

To set a suitable Cycle Time and Offset Time for thepurpose of regulating travel speed.

To introduce solution for upgrading existing traffic controlsystem which would optimize traffic flows in Kajang.

2.0 STUDY METHODOLOGY

The study consists of five main activities as shown in Figure 2.1. The

main activities are data collection, determination of phasing sequences,

determination of optimum cycle and green time split, determination of

optimum offset and development of traffic control expert system. In

addition, two activities are conducted to enhance the study output, ie.

Propose of smart surveillance system and propose of traveler information

system.


10/56

10

Figure 2.1: Study Methodology

2.1 DATA COLLECTION (TRAFFIC SURVEY)

The survey has been carried out manually at the selected intersection.

Two types of traffic count, namely, Peak Hours Junction Classified

Volumetric Count and Mid-block 16-hours Classified Volumetric Count

are required to determine optimum cycle timing and green time split plan.

In addition, Travel Time Survey and Queue Length and Delay survey are

required to determine the optimum offset.

Site Visit / Data Collection

Determination of Phasing Sequences

Determination of Optimum Cycle Time and Green Time Split

Determination of Offset

Develop Traffic Control Expert System

Proposed Smart

Surveillance System

Proposed Traveller

Information System

Proposed Automatic &

Intelligent Urban Traffic

Control


11/56

11

2.1.1 Classified Volumetric Count (at Peak Hours)

The survey is usually carried out to collect traffic data for every

directional flow at every intersection in the study focus area. Working

days at peak hours are the suitable time for the survey to be carried out.

The counts were carried out for 15 minutes duration in the morning. The

number of car in 15 minutes are multiply with four to get the total

number of car per hour.

To calculate optimum cycle time and green time splits, we used the total

number of car per hour collected previously. The data on traffic flows are

converted from classified vehicles into passenger car equivalent (pcu/hr)

by using pcu factors. In this study pcu factors is based on the study

conducted by Highway Planning Unit such adopted in Table 2.1.

Otherwise, Table 2.2 shows the traffic volume count (pcu/hr) collected

during the peak hour in the morning at the study focus areas.

Table 2.1: Adopted Passenger Car Unit (pcu) factors.

Vehicle pcu Factor

Car 1.0

Motorcycle 0.33

Van 1.0

Light Lorry 1.5

Heavy Lorry 2.5

Bus 2.0

Source: HPU, 2002


12/56

12

Table 2.2: Traffic volume in the morning at selected intersection.

Intersection -1-

Intersection -2-

Intersection -3-


13/56

13

2.1.2 Travel Time Survey

Travel speed is the main parameter for determining green time offset. In

order to determine the travel speed, the travel time survey was conducted.

During travel time survey, enumerators has used cars odometer and

stopwatch to record travelling distance and time. The average travel time

observed was 25km/hr equal to 7 m/s.

2.1.3 Queue Length & Delay

Queue length and delay was adopted to determine the queue length ofevery approach. Intersection delay study suggested by Highway Capacity

Manual 1994 has used as a guideline for conducting this survey. Average

delay of vehicles that pass through the approach and the queue length

used to compute green time offset.

2.2 PHASING SEQUENCES DETERMINATION

Phasing sequence in isolated intersection is dependant largely on

traffic demand and intersection layout. Figure 2.2 shows the phasing

sequences in the selected study focus area. For coordination purposes, the

main flows at every intersection at the arterial road under consideration

must be in the same phasing sequence. If the north direction is the main

flow, all the north bound approaches at the affected intersection are

advisable to be assigned as phase1.


14/56

14

Figure 2.2: sequence of intersection.

2.3 DETERMINATION OF OPTIMUM CYCLE TIME & OPTIMUM

GREEN TIME SPLIT

For optimum time, Webster method is used for calculation as it is a

widely used and easily understood method. The Webster formula is given

as follows (Webster &Cobbe 1966):

Where;

Co = Optimum cycle time in second.L = Lost time in one cycle which includes all red time and start up

delay.

For Malaysian condition, 3 to 4 seconds per phase can be used.

Y = Summation of critical flow ratio with saturation flows at all

approaches.

For coordination purposes, the optimum cycle time calculated using

Webster method is 120 seconds. For all signal controllers in the study

focus area are fixed to 120 seconds for first intersection,160 second forsecond intersection,and 180 second for the third intersection, the purpose

of coordination to facilitate progressive flow.

Table 2.3: The optimum cycle time and green time split at each intersections.

Intersection-1-: cycle time 110 sec.


15/56

15

Intersection-2-:cycle time 155 sec.

Intersection-3-:cycle time 175 sec.

2.4 DETERMINATION OF OFFSET

McShaneet. al. (1998), has introduce ideal offset to be

Where;

tideal = ideal offset in second

L = block length in meter

S = vehicle speed in m/s

The ideal offset is defined as the offset that will cause the specifiedobjective to be best satisfied. For the objective of minimum delay,


16/56

16

it is the offset that will cause minimum delay. More often, the ideal

offset is exactly the offset such that as the first vehicle of a platoon

just arrives at the downstream signal, the downstream signal turns

green. It is usually assumed that the platoon was moving as it went

through the upstream intersection.

Figure 2.3: A Time-Space Diagram to illustrate ideal offset.

McShaneet. al. has modified the formula by taking into account the initial

start-up delay and also the existing vehicle waiting for the green light.

The ideal offset has to be modified as follows:

Where;Q = number of vehicles queued per lane, vehicle

H = discharge hadway of queue vehicle in seconds/vehicle.

L = distance between intrsections in meter

S = speed in m/s

Loss1 = loss time associated with vehicles starting from rest at the

first downstream signal.


17/56

17

Table 2.4: The relative offsets and absolute offsets at each intersections.

Figure 2.4: Proposed signal control timing with offset.


18/56

18

3.0 PROPOSED URBAN TRAFFIC MANAGEMENT SYSTEM

The inability of the existing urban road network in the area to cope with

increased demand has been identified as one of the most pressinginfrastructure issues in the study area. Traffic problems such as

congestion, road safety deterioration, regression of mobility and

environmental effects of traffic are widely considered important

issues.Optimization of traffic control should be done in the area of

study.The main objective of the optimization is to maximize traffic flows

and minimize stopping. Past custom to counter increased congestion with

moreand wider roads, is currently giving way to more complex

management and control systems and road pricing policies. Anurban

traffic management or control system should invoke appropriate

intervening action when undesirable situations arise.

Urban Traffic Management Systems (UTMS) collect, manage and display

real-time traffic information. They are used to improve the level of

comfort and safety facilities via traffic view cameras, road information

boards and the controlling of traffic light sequences to optimize the flow

of traffic at junctions. The aim of UTMS is to develop a framework

incorporating all systems related to traffic management and traffic

control, thus creating one multi-user, multi-disciplinary traffic

management system, integrating all applications and people involved in

transportation.The required system should be intelligent and be more able

to handleactuated dynamic data, compare to existing applications.

The proposed UTMS will link together several different applications.

Within a UTMS framework several traffic management and control

applications are able to exchange data freely by using a common

specification for the storage and transfer of data. By integrating the

technology a wide range of traffic management options will become

possible. The following applications are linked up to form an intelligent

UTMS in order to solve the problems in the study area.

Traffic Control CenterAutomatic and Intelligent Traffic ControlTraffic Surveillance SystemTraveler Information System

A good communication system is very crucial in an urban traffic control

for the following purposes:


19/56

19

Synchronization of controller timer at each intersection for offsetimplementation.

Exchange of traffic data between controllers.Malfunction reporting from each controller to the control room. Incident reporting to the control room.Use of the smart camera for surveillance purpose.

Data compilation at the control room would be used for the benefit of

road users and research purposes. A wireless communication system was

selected instead of copper or fiber optic cables to avoid intrusive road

digging work.

3.1 PROPOSED TRAFFIC CONTROL CENTER

The Traffic Control Center (TCC) is the hub where all of the Kajang City

traffic control systems are monitored. The TCC is proposed to be located

at the Kajang Town Municipal Council Building and allows the various

components of traffic management (signals, control boxes, real time

video and simulations) to be effectively managed by a team of trafficengineers who monitor and maintain signals throughout the city. TCC

would serve as the main operational and control point for traffic signals.

The main function of this control center are to manage all the system and

also as monitoring base to make sure the traffic condition at its best level.

This will make sure the traffic system will get best flow and all users will

follow the system properly. TCC rely on information technologies to

connect sensors and roadside equipment, vehicle probes, cameras,

message signs, and other devices together to create an integrated view oftraffic flow and to detect accidents, dangerous weather events, or other

roadway hazards.


20/56

20

Figure 3.1: An example of Traffic Control Centre function.

3.2 PROPOSED AUTOMATIC AND INTELLIGENT URBAN

TRAFFIC CONTROL

Automatic and intelligent traffic control is a key measure in urban traffic

management, and thisis reflected in the number of urban traffic control

(UTC) systems in use today. UTC is therefore a key application within

UTMCsystems. UTC include ITS applications that focus on traffic

control devices, such as traffic signals, ramp metering, and the dynamic

(or variable) message signs on highways that provide drivers real-time

messaging about traffic or highway status.

Most of the existing urban traffic control is based on centralized control.In a centralized control system, all timings are calculated by a central

computer. The local controller would only implement the timing once it is

received from the central computer. Usually the system would consider

the traffic in terms of smoothed flow profiles. This makes the system

slow in responding to rapidly changing traffic demands, such as during

morning peak traffic growth period.

Contrary to centralized control, the proposed system is based on a fully

distributed system. In this system, all timings are calculated by the localsignal controller. Coordination with adjacent intersections is possible if


21/56


22/56

22

Figure 3.2: Expert System

The high-level module manages symbolic visual data with the goal of

assessing the current traffic situation in terms of moving, stopped, androad crossing vehicles. The module tracks all vehicles of interest in order

to identify them and compute their position, speed, and motion direction

during time. This can be used to infer real-time traffic data, which in turn

can be used for instance for traffic-light control. Tracking of individual

vehicles by vision systems is especially useful for extracting information,

such as turning rate at intersections and vehicle classification, which

could not be acquired by spot sensors.

The high-level module is designed as expert systems. Expert systemsare very often exploited in Artificial Intelligence for a large variety of

applications. They embed a large component of domain-specific

knowledge but, differently from other heuristic-based systems,

knowledge is represented in an identifiable separate part of the system

rather than being dispersed throughout the whole program.

The module architecture adopts a general-purpose model for

knowledge representation which is a production-system model with

forward chaining reasoning. Production systems process data stored in a


23/56

23

specific working memory and use data-sensitive rules rather than

sequenced instructions as the basic unit of computation. Each rule in the

production memory has a condition part to be satisfied and an action part

to be conditionally executed. Finally, an inference engine is needed for

executing rules: it matches rules satisfied by data, selects from these the

ones firing, and executes them.

The general advantages of production systems are well known: they

are easy to update and modify by adding new rules if the external

conditions change. Furthermore, they give an explicit representation of

the decision making steps with the possibility of explaining the inference

sequence to the user.

Figure 3.3: Expert System Module

3.2.2 Logical Architecture

The proposed system is based on a fully distributed system. In this

system, all timings are calculated by the local signal controller.

Coordination with adjacent intersections is possible if each controller can

provide its neighbors with some information about its status, its future

timing strategy and the time at which it expects the vehicles to leave its

intersection before the controller starts optimizing the signalizedintersection under its control. Since all timing calculations and


24/56

24

coordination are carried out at the local level, the distributed control is

able to respond almost immediately to sudden fluctuation in traffic flows.

Figure 3.4: Logical Architecture (Distributed Control Architecture)

3.2.3 Physical Architecture

Physically the system consists of three basic components, namely the

sensor which is smart camera for collecting traffic data, the controller for

controlling traffic flows at an individual intersection and coordinator for

coordinating the timing of an individual controller with its neighbors. The

Local Area Network (LAN) approach is proposed to link up all controllersas.

Each computer or microprocessor at the traffic light controller is given

an IP (Internet Protocol) address. Each computer will share traffic data and

timing with its neighbors for coordination purposes. In case where proactive

control is required such as giving priority to an emergency vehicle, the

computer at the control room will override the timing at each intersection

with pre-determined timing that gives priority flows for an intended route.


25/56

25

Figure 3.5: Local Area Network for Network of Traffic Controllers

3.2.4 Intersection Optimization

Most of the existing urban traffic control is based on a centralized control.

In a centralized control system, it calculates all timings by a central

computer. The local controller would only implement the timings once it is

received from the central computer. Usually the system would consider the

traffic in terms of smoothed flow profiles; this makes the system slow in

responding to rapidly changing traffic demands, such as during morning

peak traffic growth period. Contrary to centralized control, the proposed

system is based on a fully distributed system. In this system, all timings are

calculated by the local signal controller.

Coordination with adjacent intersections is possible if each controller can

provide its neighbors with some information about its status, its future

timing strategy and the time at which it expects the vehicles to leave its

intersection before the controller starts optimizing the signalized

intersection under its control. Since all timing calculations and co-

ordinations are carried out at the local level, the distributed control is able

to respond almost immediately to sudden fluctuation in traffic flows.


26/56

26

Figure 3.6 :The Architecture of thesystem for area Wide Urban Traffic

Control.

3.2.5 Wireless Communication System

In the present wireless digital world, it is now possible to provideintegrated communications for Kajang Councils. With their physical

infrastructure, Kajang Councils are in a very commanding position to install

digital systems with a wireless infrastructure. With the beginning of Urban

Traffic Management and Control systems (UTMC), there is a need to

provide information from the traffic light controller back to the Central

Server. Either this is an existing system, based on serial communications

which requires replacement, or a new communications connection. For

UTMC connected traffic lights, a communications path must exist back tothe control centre.

This can be accomplished by the use of Mesh4GTM Network clusters.

Traffic lights are connected via Mesh Nodes to each other, and then via an

Access Point back to the Network Control Centre. Each Mesh Node provides

an opportunity for other devices to route through it, either by direct

connection or by a wireless connection from a nearby node. CCTV cameras

for traffic monitoring can be easily and cheaply connected this way. In fact,

connections for multiple cameras can be put in place at one junction.Mesh4GTM, a street-level wireless Mesh Network, provides a low cost


27/56

27

connection between traffic lights, CCTV, variable message signs and the

Council offices. Traffic lights are converted from separate operation by the

installation of a wireless point in the control cabinet with a cable running to

the nearest traffic pole. Just a small box with an aerial is installed,

connecting to existing mountings on the pole, and installation is complete.

No major engineering is required and the traffic light can now communicate

with devices such as other traffic lights or central servers.

The wireless device can communicate through up to five or more devices

before reaching an Access Point, which then connects via a higher-level

network to the Control Centre. The higher-level network can consist of

ADSL, EPS9 circuits, 5.8GHz wireless or any other existing

communications medium. Laying copper or fiber optic cable for this purpose

is relatively very expensive and involves road digging. Renting existing

commercial telecommunication cable also involves high operating cost.

Figure 3.7: Wireless Communication System.

3.2.6 Sensor

Sensor is an essential element in an intelligent traffic control. The video

detection system is being proposed as a sensor as a replacement for loop

inductor. This system is very flexible and the price of commercial video

detection system now is reasonable and suitable with current condition.

The sensor is used either to detect the presence of vehicles or to measurethe gap or headway of the arriving vehicle in the vehicle-actuated


28/56

28

system or to count the traffic volume and to determine the queue length

in a co-ordinated adaptive system. In a more sophisticated system, the

sensor is also used to detect any traffic incident.

3.2.6.1 Smart Eye Traffic Data Sensor (TDS)

Smart Eye TDS enables monitoring of road traffic and automatic

intervention in critical traffic situations. Smart eye TDS is based on

a novel CMOS vision chip. The system determines the traffic status

from the object data supplied by the chip in the integrated signal

processor. The smart eye Centre enables simple configuration and

maintenance. The software can be installed on any Windows

PC/notebook and enables comfortable remote maintenance. Smart

eye Server enables the simple connection of smart eye sensors to

your database. Traffic data is transmitted in a format similar to

XML.

Figure 3.8:Smart Eye Traffic Data Sensor (TDS)

Smart Eye TDS operation features are as the following:

Single vehicle detection on up to 4 lanes (front fire and/or backfiredetection).

Traffic statistics for freely selectable time intervals.Warnings based on decreasing speed, vehicle numbers and time

gaps.

The prevention of traffic jams through large scale, earlydeceleration of vehicle convoys not only protects drivers nerves,

but the environments well - thanks to reduced fuel consumption.


29/56

29

The control systems require spatially distributed sensors thatcollect and reliably transmit essential information(e.g. vehicle

speed, density).

The smart eye TDS Traffic Data Sensor is distinguished by itsperformance (recording of up to 4 lanes with side or over head

mounting) and economy.

3.2.6.2 WavetronixSmartSensor HD

WavetronixSmartSensor HD is another option of smart sensor. The

WavetronixSmartSensor HD uses the latest technology to collect

consistently accurate traffic data in high definition. Patented Digital

Wave Radar II measures traffic volume, individual vehicle speed,

average speed, 85thpercentile speed, average headway, average gap,

lane occupancy, vehicle classification and presence. Operating at five

times the bandwidth, SmartSensor HD has five times the resolution of

the original SmartSensor, a detection range of 250 feet and the ability to

detect up to 10 lanes of traffic simultaneously.

SmartSensor HDs unique Dual Radar design is incredibly accurate,

providing individual vehicle speeds to within four miles per hour as well

as more precise vehicle classifications. Digital Wave Radar II reduces

spillover; works over barriers, guardrails, medians and gores; and

accurately detects partially occluded vehicles. Armed with high

definition radar, SmartSensorHD sees all vehicles in its field of view,

and not just those in pre-defined zones.

These vehicle-based detections help raise the performance bar for

SmartSensor HD. Sensor configuration is made even easier

becauseconfiguration no longer affects detection, only thereporting ofvehicles. SmartSensor HDs vehiclebaseddetection even sees lane-

changing vehicles that are often missed, or counted twice, by other radar

sensors and other technologies.SmartSensor HD is easy to install and

includesa pointing assistant for precise alignment. Like all

SmartSensors, SmartSensor HDs patented auto-configuration process is

quick and simple. HD Manage detects lanes by observing traffic flow,

and immediately provides visual confirmation of a successful

configuration. This unique auto-configuration and operation software


30/56

30

has been developed especially for Pocket PC handheld devices and

laptops.

After installation, SmartSensor HD requires little or no on-site

maintenance. Traffic data andconfiguration settings are stored in Flashmemory, so the sensor can be remotely reconfigured for optimal

performance. And SmartSensor HD is manufactured using a modern,

automated process, with surface-mounted components and integrated

antennas that provide consistent production and performance.

SmartSensor HD integrates seamlessly with existing legacy systems and

is reversecompatible with the original SmartSensor. Dual

communication ports enable SmartSensor HD to integrate with different

systems simultaneously, and flexible connectivity options make itpossible to directly retrofitSmartSensor HD into any existing radar

deployment. This, combined with high definition radar and consistent

accuracy, makes SmartSensor HD the most accurate, most cost effective

traffic monitoring solution.

Figure 3.9:Wavetronix Smart Sensor HD


31/56

31

Figure 3.10:WavetronixSmartSensor HD Operation

3.3 PROPOSED SMART SURVEILLANCE SYSTEM

The smart surveillance system consists of the smart cameras and the

microprocessors which are connected to the control room computer. The

integration of these advanced image sensors with high-performance

processors into an embedded system facilitates new application classes

such as smart cameras. Smart cameras not only capture images or video

sequences, they further perform high-level image processing such as

motion analysis and face recognition on-board and transmit the

(compressed) video data as well as the extracted video information via a

network.

Smart image sensors can overcome problems like large intensity contrasts

due to weather conditions or road lights and further blooming, which is

an inherent weakness of existing image sensors. Furthermore, noise in the

video data is reducedby the capability of video computation. Thus, the

smart camera delivers a new video quality and better video analysis

results, if it is compared to existing solutions.

The basic component of the image processing system for the smart

surveillance system comprises of the smart surveillance camera,

multiplexer, image grabber and a Windows based computer. Data input is

provided by the smart surveillance camera which produces analogue


32/56

32

electrical signals which is then digitized and stored in the frame memory

of the computer for further processing.

3.3.1 Traficon

Traficon is one of the video detection systems available in market. Based

on technology developed by the University of Louvain in 1982,

Traficon's continuous research has resulted in powerful solutions for

traffic applications. Today, Traficon is the leading reference in traffic

detection based on video image processing. Traffic managers all over the

world use its technology for vehicle detection, traffic data acquisition,

automatic incident detection, and intersection control and management in

motorway, tunnel, bridge and urban applications. As an ISO9001:2000

certified company; Traficon strives to meet its customers' requirements

by delivering reliable high quality products, tailor-made solutions and

experienced project support.

Its multi-functionality makes it the perfect traffic measurement system for

a wide range of traffic applications, including ramp metering, travel time

calculation, dynamic speed indication, queue tail monitoring, congestion

monitoring, tunnel access control, ventilation control, rerouting, VMS-

control, dynamic queue indication during road works, and dynamic lane

opening or closing.

The key factor in a Traficon detection system is the video image

processor (VIP), a standard detector board on which several types of

detection software can be run. The video signal from the camera

monitoring the traffic is used as input for the detection unit. Detection

zones are superposed onto the video image. Vehicles crossing these zones

are detected. The VIP analyses the video images to generate traffic data

and alarms. Communication interfaces link the Traficon video detection

product range with different types of communication networks: direct

line, telephone lines, fibre networks and wireless communication. On the

host computer at the control centre, the Traficon management system

(TMS PC software) monitors the video detection systems, handling

TCP/IP communication and database storage of data, alarms and images.


33/56

33

3.3.1.1 Traficon Principles

Its principles can be shown in detailed as following:

A video camera is installed at a certain height for monitoring thetraffic.

Its video signal is used as input for the detection unit. In a typical Traficon installation, a detection unit consists of a

number of VIP (Video Image Processor) boards integrated into a

standard 19" rack (together with 1 communication board).

During set-up of a VIP, detection zones are superposed on to theappropriate position in the video image.

As a vehicle crosses these detection zones thus activating them, thevehicle is being detected.

Application specific algorithms provide different types of trafficinformation

Traffic data for statistical processing Incident related data Presence data The Traficon PC Software (WATTS, TMS) monitors the

video detection system in the traffic control center.

Since video image processing highly depends on the quality of the

image received, a high quality input source is required. After all,

the camera functions as the eye of the system. Pan tilt zoom

cameras enable road operators to watch traffic and road

infrastructure from different camera positions and various angles,

wide-angle view or close-up. Traficon now offers a reliable

solution for its video detection system on PTZ cameras. Traficon's

video detection modules can store multiple configurations on one

VIP board, adapted to the traffic situation and providing accurate

and useful information on traffic data and events.

The optimal camera position depends on the type of application

(data acquisition, incident or presence detection) and the

restrictions imposed by the environment.In general, cameras should

be placed as high as possible and in the middle of the detection

zone (road - tunnel roof). If this is not possible, placement near the

fastest lane is preferred (to avoid occlusion, i.e. slow moving trucks


34/56

34

masking other vehicles on adjacent lanes). It is preferable to mount

the camera on a stable fixation (especially for long distance

applications).

The field of view depends on the height and the objective (lens) of

the camera. For detecting stopped vehicles, the detection zone is

usually limited to 350m outside and to 20 times the camera height

inside a tunnel. A camera with a high vertical inclination (i.e.

nearly looking straight down) provides a clearer distinction

between consecutive vehicles.

Figure 3.11:Traficon Smart Video Camera

3.3.1.2 Traficon Benefits

Benefits of Traficon Video Detection are as follows:

A high detection rate and a very low false alarm frequency makethe Traficon system highly reliable and a great help for traffic

operators.

The industrial set-up in a 19'' rack, with Video Image Processing(VIP) detectors, is compatible with both centralized and

decentralized detection systems.

State-of-the-art Traficon video detection algorithms perform underall weather and lighting conditions.

Open architecture makes it possible to integrate Traficon systemsinto existing traffic management systems without high costs.


35/56

35

Easy maintenance and a low overall lifetime cost.Mean Time Between Failures (MTBF) is more than 20 years for all

Traficon equipment.

Easy to install, easy to adjust to changing traffic situations, easy toextend and easy to update to additional traffic requirements.

Incident detection and alarms can be fine-tuned to meet customizedapplication requirements.

Faster detection means a faster reaction and a better chance ofpreventing secondary incidents

3.3.1.3 Traficon Applications

Traficon, which makes the core hardware and software for

detection system in industrial grade modular processor card

formats, is increasingly emphasising whole solutions for customers

rather than simply products. Because the Traficon video detection

system is multi-functional, fast, flexible and reliable, it is the

perfect traffic measurement system for a wide range of traffic

applications such as:

Vehicle presence detection Incident detectionCertain circumstances detection ( movement of ambulance, patrol

cars and fire engines)

Vehicles classificationsVehicles speed measurementTraffic countingQueue lengthFigure 3.12 shows the detection zones in the sensors filed-of-view

zone in each lane. Vehicles detected and tracked within the tracking

zone in each lane. The spatial signature of each detected vehicle is

integrated with the temporal signature of its motion obtained from the

vehicle tracker to measure speed of the detected vehicle. This

information is used in obtaining the measurements shown in Figure

3.13. Tracking and speed measurement help in the treatment of

artifacts such as shadows and perspective interference of


36/56

36

neighboringvehicles (perspective interference makes the vehicles

sometimes appear to be moving in two adjacent lanes).

Figure 3.12: Integrated video sensor tracks vehicles and uses speed for

enhanced accuracy.

Figure 3.13: Integrated video sensor provides a large number of trafficparameters for traffic management and control.


37/56

37

Figures 3.14 and Figure 3.15 show tracking of moving vehicles and

the detection of the vehicles stopped in a queue at an intersection.

The queue measurement software detects and tracks the position of

vehicles within each tracking detector. In the process of detecting

and tracking vehicles, it also calculates the vehicles speeds and

lengths; and this, in turn, enables the calculation of the size of the

queue, the number of vehicle stops, the percentage of roadway

covered by vehicles, and the number of vehicles entering and

exiting the detector. Based on the way the detector detects and

tracks vehicles, the following MOEs can be measured:

1) Queue Size: The number of stopped vehicles (a stoppedvehicle is defined as one traveling less than 5 mph).

2) Queue Length: The distance from the first stopped vehicle tothe end of the last stopped vehicle.

3) Queue Exit Speed: The average speed of vehicles exiting thedetector during an interval.

4) Queue Exit Volume: The number of vehicles exiting thedetector during an interval.

5) Queue Flow Length: The cumulative length of vehiclesexiting during an interval.

6) Total Stops: The total number of stops made by vehicles thathave exited the detector during an interval.

7) Spatial Occupancy: The percentage of length of the roadwaythat is occupied by vehicles.

Figure 3.14: Moving vehicle being tracked with red overlay in the display.


38/56


39/56

39

Figure 3.16: AIDAIncident detection detector configuration.

The flash flood might be occurred if improper drainage system is

designed. The detection of this disaster also can be detected by the smart

surveillance system. By detecting this disaster earlier, the loss can be

minimize. The disaster can be detect by the video detection cameras

which are located at each intersection. Then, the information about the

flash flood during the heavy rain will be submitted to the Traffic Control

Centre. The Control Centre will take an action informing the user at

Kajang Town.

This system will be integrate with SCADA real time rain fall and river

water level by Drainage & Irrigation Department. When there are heavy

rain that can caused flash flood an alert system will trigger the sirens/

alarms and information also will be displayed at VMS board. At the same

time this information will be conveyed to the responsible authorities to

make sure they are in alert conditions.


40/56

40

Figure 3.17: The Disaster Detection & Responds System.

3.4 PROPOSED TRAVELLER INFORMATION SYSTEM

The traveller information system consists of the Traffic and Traveller

Information responding to each media (communication medium, etc.). It

collects and provides the information related to the Traffic and Traveller

Information responding to each media.

The traveller Information System assist user to plan their journey by

helps them in making decision on the best route to the destination. The

data collected from current flow will be analysed. Then, it will give the

user real-time information about the route condition. The system also can

provide the traffic estimation and prediction to certain selected route. The

user canexperience the broadcast and interactive traveller information.

They can access to the information in various of way.

List of Traveller Information System support as stated below;-

Variable Message Sign (VMS)

Internet AccessCar Navigation System

traffic incident / traffic

breakdown / flood

trigger alarm alert(emergency light)

VMS information

SMS alert to localautority, police,

JKRetc

manual control

system


41/56

41

Call CentreRadio/TV StationSMS/MMS System or Personal Digital Assistant

3.4.1 Variable Message Sign

A variable message sign is an electronic sign often used on roadway to

give travellers information about special events. Such signs warn of

traffic congestion, accidents, roadwork ones, or speed limits on a specific

highway and road segment.

Also provide the user to take alternative routes, limit travel speed, warn

of duration and location of the incidents or just inform the traffic

conditions.

A complete message on a panel generally includes a problem statement

indicating incident, roadwork, stalled vehicle, a location statement

indicating where the incident located, an effect statement indicating lane

closure, delay and action statement giving suggestion what to do traffic

conditions ahead.

Figure 3.18: Variable Message Sign

Some Specifications available in the market are:

2-line signs offer 12 characters of 320mm text, or 16characters at 400mm

3-line signs offer 18 characters at 400mmAluminium enclosure, fully welded with internal strength

and support members

Dual colour upper and lower flashing red and amber lanternsalso use long life

Operating temperature range -20C to +60C


42/56

42

Window material: 6mm anti-reflective, UV-protected,polycarbonate with anti scratch coating

Optical performance conforming to UK Standard TR2136/European Standard EN 12966

Visibility of 400mm characters > 300mBrightness controlled by ambient light over 16 levelsDisplay angle 5 horizontal and verticalCommunications: NMCS2 Protocol over RS485 bus, with

GSM, radio, TCP- P,

modem or Internet options possible

Location Proposal for VMS at 6 locations

1. From Seksyen 8 Bangi to Kajang (Before SILK Highway

Intersection)

2. From SILK Highway to Jalan Bukit (KTM Commuter)


43/56

43

3. From JalanReko to Kajang before SILK Highway

4. From Sungai Ramal to Kajang at Kenari


44/56

44

5. From Cheras to Kajang at Grandsaga Highway before Sungai

Sekamat Interchange


45/56

45

Figure 3.19: Proposed location for VMS.

3.4.2 Internet Access

In Kajang Municipal area, Municipal Council and JKR (together get the

budget) need to extend the webpage facilities to enable the traffic

information updated hourly or more frequent to their existing webpage.

The webpage also can provides information on incidents, reports of road

closures, road works or any events impacting on traffic flows in the

Kajang City.


46/56

46

Figure 3.20: Existing Webpage for Kajang Municipal Council

Figure 3.21: Example, life traffic webpage


47/56

47

Figure 3.22: Example; Travel time in life Traffic Webpage

3.4.3 Car Navigation System

This approach was implemented widely in develop country such as Japan,

UK and USA. But in Malaysia recently the Navigation system become

popular to the car user, nevertheless the system is not interacting with the

life traffic at the road. So at this time the navigation system only guide

the user to the destination either faster, nearest, toll highway and non toll

highway. The new technology in this system, allow the system to link to

the traffic website. From there, they can access the information about the

traffic condition and guide the user to the smooth route.

Figure 3.23: Car Navigation System (GPS)


48/56

48

3.4.4 Call Centre

Kajang Municipal Council and JKR suggested to set-up one medium size

call centre or control centre to provide information and control the trafficin Kajang area especially at Kajang City Centre. It can provide automated

half-hourly to hourly reports of traffic situation as well as ad hoc

reporting of major situation in Kajang City. Traffic information on main

route is also automatically updated to the interactive voice response

system. If the traveller require more personal response call agents at the

Call Centre will also be on hand to answer specific queries within the

Kajang city.

Figure 3.24: CallCentre and Traffic Monitoring room

3.4.5 Radio / TV Station

Traveller can also get the information of the current traffic condition from

the radio and TV station. Nowadays, many radio and TV station concern

about their listener and viewer demand that wish to know about the traffic

condition at the morning and evening peak hours. So, the radio and TV

can collaborate with Traffic Control Centre for the Kajang City as

example below:-


49/56

49

Figure 3.25: Radio and Television traffic report update

3.4.6 SMS/MMS System or Personal Digital Assistant

By cooperating with the telecommunication service provider, the trafficcondition can be attained by the communication tools which are hand

phone and personal digital assistant (PDA). The traveller can send SMS

or MMS to get the information about traffic flow. They also can

subscribe the information about traffic flow at the peak hours for monthly

payment from the telecommunication service provider.


50/56

50

Figure 3.26: SMS/MMS and PDA traffic update

4.0 OUTCOME

The outcomes of this study are outline as below.

4.1 LOW COST SOLUTION

Low cost solutions are the outcome of this study, ranging from setting the

optimum timing manually to an intelligent system with communication

system. The intelligent system is based on distributed control system

using microprocessors whereas the communication system is based on

wireless system or system using power cable as the communicationmedium to minimize cost.


51/56

51

4.2 PROPOSED COST OF INSTALLATION

In this section, we will list out the typical costing for installation of all the

four intersection with ITS.

Table 4.1: Cost for Setting up the system.

4.3 Calculating Congestion Costs and Reduction Benefits

Various methods are used to quantify congestion costs. The most

appropriate approach for many applications, although difficult to perform,

is to calculate the marginal delay caused by an additional vehicle enteringthe traffic stream, taking into account the speed-flow relationship of each

road segment. Another approach is to determine the user fee needed to

reduce demand to design capacity, based on travelers willingness-to-pay

for road use. A third approach is to calculate unit costs of currentexpenditures on congestion reduction projects. In theory these three

methods should produce similar values, assuming that roadway capacity

is expanded based on vehicle delay costs as reflected in vehicle users

willingness to pay, but in practice they often provide different results. In

addition, necessary data is often limited, making accurate congestioncosting difficult.


52/56

52

In this case, we considered the value of time as main contribution to the

cost benefit for this congestion cost calculation (Lowest case scenario).

Table 4.2: Cost for Congestion

As we can see in the above calculations, the cost of congestion is

RM6.72 million/year compare to setting up of ITS which is RM 2.07

million (including operation and maintenance cost). So we strongly

advise this KUTMS (Kajang Urban Transport Management System) needto implement very soon to avoid the price increasing and avoid the stress

and waste of time for the user and people in Kajang area.

4.4 SITE INSTALLATION

The traffic operations centre is equipped with 2 closed circuit televisionmonitors, which are connected to on-street cameras located on the arterialroutes. The cameras assist the traffic staff in recognizing traffic problems

and making remote timing changes via the interconnected traffic signals

to help in maximizing the efficiency of the on-street traffic flows. During

the design phase of the project, communications were the primary focus

and various technologies were researched.

In order to manage ITS system, we propose to use wirelesscommunication system. It was considered the most cost effective and


53/56

53

technologically advantageous option considering existing needs, future

expansion needs and incorporation of other Intelligent Transportation

Systems (ITS) traffic management components such as variable message

signs, real time street conditions based on street sensor feedback andintegration into a regional traffic management centre. Many othertransportation related uses have been realized through the use of visual

information relayed from closed circuit television. Some of these uses are

listed as follows:

Visual verification of reported signal malfunctions-allowsverification

of malfunctions prior to dispatching signal crews.Visual verification and identification of congestion reported via

system sensors.

Incident Detection - allows for quick and correct dispatch ofemergency personnel and provides basic building block for later

incorporation of incident management system.

Allows for remote traffic counting capability - Turning Movement,Volume, Occupancy and other traffic information can be collected

from the central control room deleting the necessity to field locate

personnel

Delayed traffic counting capabilities - Allows for recording oftraffic conditions, which can be tabulated at a later date as time

permits.

Identification and/or confirmation of missing traffic controldevices.

Evaluation of needed improvements to pavement and pavementmarkings.

Roadway drainage deficiencies.Time savings provided by the ability to view more than one location at

one time. In conjunction with the computerized signal interface, the

traffic operation centre is equipped with 2 closed circuit televisionmonitors, which are connected to on-street traffic camera located on the

arterial.


54/56

54

Figure 4.1 : Installation With Local Network

They comprise of four video cameras, an industrial PC, an image

grabbing card, a multiplexer and support equipment such as video

recorder and uninterrupted power supply which were placed beside thetraffic light controller. Below is Figure 4.2 showing the camera as a

sensor and using existing steel pole that can be maximized for installation

of cameras.

Figure 4.2: Camera with sensor.


55/56

55

5.0 CONCLUSION

Most of the existing traffic controllers are based on vehicle actuated

system. This system is good for isolated intersection if the maximum time

of each phase is calculated to optimize traffic flows. However the systemcould not be coordinated to optimize a group of traffic controllers

because its actions are unpredictable. Multi-plan timing system could beset based on computed timing that gives optimum traffic flows.

The optimization procedure includes individual controller setting and

offset timing to minimize stopping and maximize flows that give

progressive flows with maximum bandwidth. Optimizing existing traffic

controllers are relatively easy although it requires surveys and on site

setting. However to maintain optimum operation, constant monitoring is

needed especially if power supply is not stable. It is undeniable thatsetting up a ITS system in this Kajang town area would be a very wise

decision as it will help to lessen the congestion in Kajang and also will

benefits the town here as has smoother traffic flow.

5.1 RECOMMENDATIONS

Initial and maintenance works to optimize existing traffic controllers

consume a great deal of time and energy. If this operation can be

automated intelligently, the traffic flows could be optimized in real time

automatically. For this reason, the study team recommends that:

Upgrade the existing controllers to controllers withmicroprocessors

Install advanced sensors Install communication system to facilitate data exchanges between

traffic controllers which are necessary in optimizing traffic flows.

Other recommendation if the congestion still cannot be manage by this

system are stated below: Introduce road pricing within peak hour in the City centre as

Figure 5.1

Road pricing also can be use for the operation and maintenance for the

ITS system to reduce the burden from the government. But the public

transport system and service must be upgrades prior to this road pricing

implementation.


56/56

Figure 5.1 Road Pricing implementation

Introduce Response team during peak hour by Municipal Councilor JKR to control the traffic during accident and incident happen at

site and as part of the reported team from site to Control centre or

caller room as show on Figure 5.2

Intelligent Project (Final)

Documents