Design of Dynamic Traffic Signal Control System

The research article under the title “Design of Dynamic Traffic Signal Control System” has been written by a team of four people comprising Shabnam Sayyed, Prajakta Date, Richa Gautam, Gayatri Bhandari where all of them are hailing from Department of Computer Engineering of Bhivarabai Sawant Institute of Technology & Research, India. In the introduction section, the authors have successfully highlighted the consequences of traffic congestion. They have included several alternative methods, which are currently in use for controlling the traffic congestion, but they have clearly identified the significant impact that a dynamic control of traffic can have on the reduction of traffic congestion.

Few of the major drawbacks of the presently used traffic control systems have been pointed out by the research article team. They are,

• Heavy traffic jams
• Having to wait even if there is no traffic
• Emergency Car Problems (priority is not given for emergency vehicles)
• Inadequate traffic information to users

The research article consists of a literature review section, which provides a clear explanation on the methods that have been tested in the past.

Expert System

The process related to expert systems is that it uses a set of rules to decide the next action in traffic light signal system. The actions specified here are capable of changing some of the control parameters, resulting in a newly implemented system. The system is also said to have the potential to learn new rules. The final outcome of the solution is that the researchers have identified that the system performance could be improved, but some simplifying assumptions had to be made in order to avoid too much of computation.

Prediction Based Optimization

This particular model had been introduced by Tavladakis and Voulgaris in 1999. The measurements taken during the current cycle are utilized to test possible settings for the next cycle. The system uses data of a single cycle and therefore, it is not capable of handling strong fluctuations in the traffic flow. Although the system has seemed to be a highly adaptive, the rate at which it adapts has been too quick. This has resulted in poor performance of the system.

Fuzzy Logic

The fuzzy logic controller has shown more flexibility than the fixed controllers and vehicle actuated controllers. This phenomenon allows the traffic to flow in a smoother manner to reduce waiting times. However, the disadvantage that had been identified is its dependence on the preset qualification values for the fuzzy variables.

Reinforcement Learning

The reinforcement learning is a system which is known to have first been studied by a person called Thorpe. He uses a traffic light based value function where a neural network is used as the value function. The task of value function is to predict the waiting for all cars that are standing at the junction.

        Magneto – Resistive Sensors

  Cai Bai-gen et.al. (2009) has designed a system for detecting vehicles which is based      the usage of magneto – resistive sensors composed by wireless traffic collection nodes.  The nodes are set two sides of road to detect vehicles signals. According to the research  article authors, the magneto – resistive sensors are costly and in cases where the system  fails, the maintenance cost too will be higher.

The proposed system by the research team have several objectives where the primary objective for them is to come up with a user interface and a hardware implementation of the dynamic traffic control system. In addition to that, the system should have an efficient and a safe traffic flow which reduces the waiting times at junctions. The fundamental idea of them is that traffic congestions can be reduced by increasing the green signal duration on busy roads while decreasing the red signal duration on non-busy roads.

The main component used for the project are IR-LED transmitters and receivers and those resources are utilized to measure the traffic flow. Infrared sensors are mounted on road to detect the frequency of the vehicles. The proposed system is known as Dynamic Traffic Light Control (DTLC) and it senses the presence or absence of a vehicle through the sensor assembly mounted on each road where it acts as an input to the DTLC unit. This particular input signal indicates the vehicle density on each road. The basic operations of the system are implemented by using a Microcontroller 89c51AT. The DTLC system then generates the required output signals for Red, Green and Yellow signal lights while monitoring their respective time durations. It also considers the length of the vehicles on each road, in this process. The system structure is further depicted in the figure shown below.

Figure 01 - Basic Structure of the Proposed System

The proposed system in the research article provides a very good explanation as well as a foundation for us to build our system. The article also brings up a very clear and specific literature review where it highlights the advantages as well as disadvantages in each of the previously used methods to overcome the traffic congestion problem.