Model Of Double Isolated Intersection Computer Science Essay

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

In this section, overall description of method that will be used in this study is elaborated. This study is focusing on developing the intelligent fuzzy traffic controller for double isolated intersection. Mamdani-type fuzzy inference system (FIS) is using as editor to develop fuzzy rules, input and output membership functions. MATLAB program will be used in implementing the whole project. In this project, graphical user interface (GUI) for fuzzy based modeling using GUIDE and Fuzzy Toolbox in MATLAB will be developing. The isolated intersection traffic model is design using SIMULINK block diagram. In this project, the Multi-phased fuzzy traffic controller is proposed with consists of Fuzzy Green Phase Extender and Fuzzy Phase Selector. The rules of both are different. At last, the performance of proposed controller will evaluate by simulation as compare to actuated traffic signal controller.

3.1 Model of Double-Isolated Intersection

The proposed traffic signals for double-isolated intersection shown in Figure 13 are designed based on Mustafa (2009). In this model several assumptions has been made. The distance between two junctions, D is less than 50 meter, and no vehicles are stop or present at this distance at any time. In the other words, the green light must actuate until there is no vehicles at the link between two intersections. It can achieve by placing the detectors to detect the present of vehicles between the intersections. The other assumptions are, only one phase will be triggering the green light at one time. For example, if phase 1 is actuated (green light is ON) allowing vehicles from West go through to 1st and 2nd intersections, the others phase (phase 2, 3, 4, 5 and 6) are not actuated (Red light is ON) and so on.

Figure : Propose Double-isolated Intersection Model and Location of the sensors based on Mustaffa (2009)

The proposed intelligent fuzzy controller allow the vehicles to go straight, turn right or turn left at any intersection when the green light is ON at corresponding lane or intersection. There are two sensors placed on the road for each lane. The first sensor behind each traffic light counts the number of vehicles passing the traffic lights, and the second sensor which is located behind the first sensor counts the number of vehicles coming to the intersection at distance L from the lights. The number of vehicles between the traffic lights is determined by the difference of the reading between the two sensors. The distance between the two sensors L, is determined accordingly following the traffic flow pattern at the particular intersection. The fuzzy logic controller is responsible for controlling the length of the green light time and the phase sequences according to the traffic conditions.

3.2 Proposed Design of Intelligent Fuzzy Traffic Controller

The fuzzy traffic signal controller for this project is designed using Mamdani-Type fuzzy inference system in MATLAB Toolbox. Currently, there are two types of fuzzy rules namely, Mamdani fuzzy rules and Tagaki-Sugeno-Kang (TSK) fuzzy rules. Mamdani's controller are using min-max operator on fuzzy rules to make a decision. The main difference between the two methods lies in the consequent of fuzzy rules. Mamdani fuzzy systems use fuzzy sets as a rule consequent whereas TSK fuzzy systems employ linear functions of input variables as rule consequent. All the existing results on fuzzy systems as universal approximations deal with Mamdani fuzzy systems only and no result is available for TSK fuzzy systems with linear rule consequent (Sivanandam, 2007). The most importance reasons to choose Mamdani FIS method because it is the most commonly method and defined it for the Fuzzy Logic Toolbox. It has widespread acceptance, enhances the efficiency of defuzzification process and well suited to human input (Mustafa, 2009).

The proposed Intelligent Fuzzy Traffic Controller consists of two parts. The first part is (i) Fuzzy Phase Selector and second part is (ii) Fuzzy Green Phase Extender, based on Zarandi et al. (2008), Azura et al. (2010) and Murat et al. (2005) work. In this fuzzy controller, fuzzy phase selector function and fuzzy green phase extender function are located in different level of the multi-level signal control system as presented in Figure 14.

Fuzzy Green Phase Extender

Extent or terminate the current green phase

Fuzzy Phase Selector

Selecting the next green phase

Intersection

Detector Data

Extend the Current Green Phase

Terminate

Next Green Phase

Figure : Multi-phase Fuzzy Traffic Controller Block Diagram (Zarandi et al. 2008)

3.2.1 Fuzzy Green Extender Function

Green light extension time of the green phase is produce by using this function according to the condition of observed traffic flows. The traffic flows are observed by receives data from the detector. The two detectors are located in each arm. The extender will determines the right timing for the green phase by tuning the duration of the current green phase with green extension of different lengths, or by terminating the current phase. The fuzzy rules compare traffic conditions with the current green phase and traffic conditions with the next candidate green phase.

3.2.1.1 Input Parameters of Fuzzy Green Extender Function.

Input parameters of fuzzy logic green extender are the following:

(a) The Longest of the Queues in the Red Signal (QR)

This parameter determines the approach of an intersection which has longest queues during the red signal. The input membership function of QR will subdivided into five ranges: Very Short (VS), Short (S), Long (L), Very Long (VL) and Extremely Long (EL). Each range will correspond to a membership functions.

(b) The Number of Vehicles Approaching the Green Signal (QG)

Approaching vehicles during green signal is selected as the second input parameter of the green extender functions. The input membership function of QG will subdivided into five ranges: Very Few (VF), Few (F), Moderate (MD), Many (M) and Too Many (TM).

(c) The Present of Vehicles in Link of Intersection (PV)

The present of vehicles at the link between intersection 1 and 2 is selected as third input. The membership functions are: Present (P) and Not Present (NP)

These two input parameters are considered simultaneously while controlling the traffic flow in this proposed traffic controller. The membership functions of all parameters will be determines using Gaussian Membership Functions.

3.2.1.2 Output Parameters of Fuzzy Green Extender Function.

Output parameter of fuzzy logic green extender is following:

(a) Extension Times (ET)

Extension time which means the extension time of green lights at the current green phase. It will subdivide into five ranges of membership functions of ET: Zero or Terminate (Z), Short (S), Long (L), Very Long (VL) and Extremely Long (EL). The membership functions of all parameters will be determines using Gaussian Membership Functions

The example of rules of Fuzzy Green Extender Function can be expressed as:

IF Queues in the Red Signal (QR) is Very Short (VS) AND

Vehicles Approaching the Green Signal (QG) is Very Few (VF) AND

Vehicles at the link (PV) is Not Present (NP)

THEN Extension Time (ET) is Zero or Terminate (Z) the current green phase.

3.2.2 Fuzzy Logic Phase Selector

The phase selector determines the most suitable phase order. Phase selector controls the phase sequence based on the vehicle's queue length and the extension time of green light from Green Extender Function. This is accomplishing by selecting the next green phase. The traffic situation will be monitor continuously based on data from the detectors and when the green phase is terminated, the decision of the next phase is updated (Zarandi et al., 2009). Figure 15 present the phases and the basic phase order of the intersection model (Figure 13). For example, if the current green phase 1 is to be terminated, the phase selector decides whether to launch next the phase 2 or phase 3 or phase 4 or phase 5 or phase 6. It means, the normal cycle 1 - 2 - 3 - 4 - 5 - 6 can be changed for example 1 - 2 - 4 - 3 - 4 - 6 - 5 - 1, depending on the traffic situation. In this system there is freedom in phase sequencing.

3.2.2.1 Input Parameter of Fuzzy Phase Selector

Input parameter of fuzzy phase selector is following:

(a) The Longest of the Queues in the Red Signal (QR)

This parameter is the most deterministic parameter in fuzzy logic phase selector. This is used for determining the phase sequence in fuzzy logic phase selector. The fuzzy inference is based on weights of each phase. The weights can be defined by the number of queuing vehicles waiting for the green signal in each red phase. Each phase will have its own membership function. The rules are formed to give priority to the phase with highest demand for green time.

For example, if the phase 1 is just terminated the phase selector rules based on number of vehicles in the red signal for another 5 phase (2, 3, 4, 5 and 6). . The input membership function will subdivided into five ranges: Very Short (VS), Short (S), Long (L), Very Long (VL) and Extremely Long (EL)

Phase 1

Phase 4

Phase 5

Phase 6

Phase 3

Phase 2

Figure : Basic/Planning Sequences of the Signal Control at the Intersection Model

3.2.2.2 Output Parameter for Fuzzy Phase Selector

Output parameter of fuzzy phase selector is following:

(a) Phase Ordering (PO)

Phase ordering is the output parameter of the fuzzy phase selector for this proposed controller. According to the input parameter, the next phase will be selected and a proper phase sequence will be decided.

The example of rules of Fuzzy Phase Selector Function can be expressed as:

Assumed, Green Signal for Phase 1 is just terminated, and then fuzzy phase selector will determines the next phase that should be trigger using this rules. For example:

IF Queues in the Red Signal (QR) on lane 2 Very Long (VL) AND

Queues in the Red Signal (QR) on lane 3 Very Short (VS) AND

Queues in the Red Signal (QR) on lane 4 Long (L) AND

Queues in the Red Signal (QR) on lane 5 Very Long (VL) AND

Queues in the Red Signal (QR) on lane 6 Short (S)

THEN lane 2 is chosen as next green phase.

3.2.3 Rule bases of the Proposed Traffic Controller

There are two rules bases on this controller. One of them is relevant with the fuzzy logic green extender and the other is the fuzzy phase selector. Rule bases will be built on the combination of inputs and output parameters. Mamdani method will be used for developing the rule bases where using this method the consequent part is defined as fuzzy sets as antecedents part of the rules

3.2.4 Inference and Defuzzification of Proposed Fuzzy Traffic Controller

Mamdani (max-min) Inference Method will be used in this controller. The defuzzification process will be applied to get the crisp value. Defuzzification means conversion of the fuzzy values to the crisp values. The method of defuzzification will be used either Center of Gravity (COG) or Mean of Maximum (Sivanandam et al., 2007).

3.2.5 SIMULINK, SimEvent, Graphical User Interface (GUI) and Simulation.

The isolated traffic intersection model will develop in MATLAB using SIMULINK and SimEvent toolbox. The Graphical User Interface (GUI) will also develop using GUIDE (Graphic User Interface Developing Environment) in MATLAB. The purpose to develop a programmed GUI is to interact with fuzzy variables in order to model the traffic controller with different inputs.

The actuated traffic signal controller for double isolated intersection will also develop in this study in order to compare their performance with proposed fuzzy traffic controller. Simulation method will be used to test the performance of fuzzy traffic controller and the result are discussed based on (i) average waiting time, (ii) average delay time (deceleration, stop, acceleration delays) and (iii) average queue lengths as performance index for controlling traffic flow at the intersection. Table 3 present the example table that will be used to compare the performance is of Fuzzy Traffic Controller and Actuated Traffic Controller. The values of each parameter are gain from the simulation on different conditions of traffic volumes such as Low Volume, High Volume and Saturated Volumes of traffics.

Table 3: Performance of Fuzzy Traffic Controller AND Actuated Traffic Controller ( Azura Che Soh et al. 2010)

Performance

Measure

Phase

Fuzzy Traffic Controller

Actuated Traffic Controller

Improvement

(%)

average waiting time

(seconds)

Phase 1

:

Phase 6

average delay time

(seconds)

Phase 1

:

Phase 6

average queue lengths

(seconds)

Phase 1

Phase 6