CHAPTER ONE
INTRODUCTION
1.1 Background of the study
The monitoring and control of vehicular traffic and pedestrians pose a major challenge to transport authorities around the world. The escalating number of vehicles in cities not only has a huge environmental impact, but also results in loss of lives on the road. This situation demands a comprehensive approach involving a system in which both the traffic controls for vehicles and pedestrians are coordinated so that road users are safe and traffic is smoothly flowing. Currently, pedestrian crossings pose a significant hazard in many countries, both in developed and developing countries due to the increase in the number of vehicles. Each year a staggering figure of 500,000 pedestrians are killed all over the world and in China alone from 2000-2004, half a million pedestrians were killed (Zhen Liu, Simulation of Pedestrians in Computer Animation in Proceedings of ICICIC (2) 2006. pp. 229~232.).
The European Transport Safety Council (ETSC) claims that 15 to 30 percent of the transportation mode used is walking. According to a telephone survey conducted by the Royal Automobile Club of Spain in the year 2000, walking is highly recommended as part of a healthy lifestyle with no negative side effects. However, it has been the victim of badly controlled traffic, thus increasing the mortality rates of road users. In the large cities of Europe, especially in Spain, people walked to their destinations but this is being seen as dangerous as pedestrians are more vulnerable to road accidents than passengers and drivers of cars (European Transport Safety Council. In a conventional traffic light controller, the traffic lights change at a constant cycle time which is clearly not the optimal solution. The system calculates the cycle time based on average traffic load and disregards the dynamic nature of the traffic load, which aggravates the problem of congestion.
Consequently, we see an urgent need to optimize traffic control algorithms to accommodate the increase in vehicles in urban traffic that experience long travel times due to inefficient traffic light controls and to improve pedestrian’s safety.
In this paper, we propose an optimal control of traffic lights using a genetic algorithm (GA), in a four-way, two-lane junction with a pedestrian crossing. The innovative design of the pedestrian crossing is also based on such algorithm, which includes pedestrians as one of the parameters. The specific genetic algorithm used in this work is a standard genetic algorithm. A Genetic Algorithm is an adaptive and efficient heuristics that is able to solve optimization problems. This is a stochastic search technique to look for optimal solution. Most genetic algorithms are used in research and science related works to look for optimal solutions. They usually run on powerful computers as genetic algorithms generally are resources taking in terms of CPU time and memory size. Some methods a genetic algorithm uses are selection, crossover and mutation inspired from evolution in the real nature. Genetic algorithm is introduced in the traffic control system to provide an intelligent green interval response based on dynamic traffic load inputs, thereby overcoming the inefficiencies of conventional traffic controllers. In this way, the challenges are resolved as the numbers of vehicles are read from sensors put at every lane in a four-way, two-lane junction and pedestrians are monitored at the road junction.
The features inherent in genetic algorithm play a critical role in making them the best choice for practical applications, namely optimization, computer aided design, scheduling, economics and game theory. It is also selected because it does not require the presence of supervisor or observer.
However, genetic algorithms, without prior training, continuously allow permanent renewal of decisions in generating solutions. Instead of trying to optimize a single solution, they work with a population of candidate solutions that are encoded as chromosomes. Within these chromosomes are separate genes that represent the independent variables for the problem at hand.
There are a number of specific attributes of genetic algorithms that give them an edge over other traditional optimization techniques. These are:
1. A genetic algorithm works from a population, not a single point, and hence it is likely to be trapped at a local optimum.
2. Derivative freeness, i.e., a genetic algorithm does not need the objective function’s derivative to do its work.
3. Flexibility, i.e., a genetic algorithm can function just fine regardless of how complex the objective function is; the only thing it requires of the function is that it is executable (i.e., its value can be calculated given the values of the decision variables).
4. Because of its implicit parallelism, a genetic algorithm can handle combinatorial problems efficiently. It has been shown that as the size of the search space or number of solutions increases exponentially, the time requirements for the genetic algorithm to reach a solution only grows linearly. This feature is particularly useful for on-line optimization of transportation problems such as traffic control.
5. A genetic algorithm naturally lends itself to parallel implementation. This follows from its functional components structure.
6. Genetic algorithm is, for the most part, based on intuitive notions and concepts.
The preliminary review of the literature indicates that genetic algorithm has not been tested on pedestrian crossings. This work has, therefore, attempted to implement this algorithm and study its effects on this problem.Ayad Mashaan Turky, Mohd Sharifuddin Ahmad and Mohd Zaliman Mohd Yusoff, Use of Genetic algorithm for Traffic Light and Pedestrian Crossing Control, (2009). pp.1-2.
From a recent analytical statistics of the Nigerian Ministry of Transportation 2010, it is estimated that roughly half of the congestion is what is known as recurring congestion, which is caused by recurring demands that exist virtually every day, where roads use exceeds existing capacity and bad roads. The other half is due to non-recurring congestion caused by temporary disruptions. Four main reasons of non-recurring congestion are: traffic incidents (ranging from disabled vehicles to major crashes), work zones, weather and special events. Expert systems with Application systems dramatically reduce available capacity and reliability of the entire transportation system. Therefore, researchers have carried out many researches to increase capacity and remove bottlenecks. Schaefer, Upchurch and Ashur (2008) developed a simulation model for evaluating freeway lane control signing. The simulation results show that the lane control has little influence on congestion. However, the region between heavy and medium traffic flow is sensitive to lane control. This is why the Genetic algorithm has been proposed to solve this major problem, because by making use of this algorithm, traffic quality can be improved and operation costs can be reduced.
1.2 Statement of the Problem
Port Harcourt and indeed Nigeria alongside other developing countries are facing serious traffic congestion problem due to rapid motorization and rapid population growth in their cities. Infrastructure development could not match the rapid motorization. As a result, serious congestion occurs almost at every intersection during peak hours mainly because of the inability of signal system to provide optimum flows, either due to the imbalance green time split or optimum band width for progressive flows. Thus this project work has come with the view of designing an expert system to check this difficulty in the urban and sub-urban metropolitan areas of our country.
1.3 Objectives of the Study
The aim of this project is to develop and design an effective traffic control system that can monitor and regulate traffic congestion within the road network in our cities. Aims include:
1. To design a traffic system that is not only time based but traffic based on highly packed junctions.
2. To design a traffic control system that will be available round-the-clock
3. To design a system that coordinates traffic flow using the appropriate programming language.
1.4 Significance of the Study
The uniqueness of the project is not only for clearing the traffic but it shares time slot equally between two sides of a junction. It is going to be a useful project for traffic police to prevent road accidents and promote safety on the road for road users. Since automotive technologies are gaining ground in modern day traffic-control systems and the number of vehicles and passengers is rapidly growing, traffic control systems are needed to ensure the safety of all parties involved which include, pedestrians who obediently wait for traffic signal to interrupt traffic so they can cross and drivers who also patiently wait for their turn to move.
1.5 Scope of the Study
This work is essentially intended to design an efficient traffic control system to control traffic in the urban areas and industrial estates across the country. It looks at controlling traffic in a four-way, two lane junction.
1.6 Limitations of the Study
This study was limited by some factors which include:
1. Limited materials: due to some materials being inaccessible, I only had to work with the few I could get.
2. Technical issues: owing to the fact that my system’s battery had expired, I could only work on my project when there was power supply.
3. Time factor: because of the challenges mentioned above and others, the time given to finish my project work was not enough.
1.7 Definition of Terms
It’s pertinent to highlight and define properly some important terms that are used in this work. They include:
Algorithm: a process or set of rules to be followed in calculations or other problem-solving operations, especially by a computer.
Congestion: the state of being overcrowded, especially with traffic or people.
Control: this is the action necessary to ensure that plans and objectives are being achieved. Control as applied to road traffics is the act of directing vehicles (motorists) and pedestrian traffic around a construction zone in order to be free from accidents and other road disruptions.
Genetic: of or relating to genes or origin.
Traffic: this is defined as the vehicles that are travelling in an area at a particular time.
Traffic light: which is also known as ‘stop light, traffic lamps, traffic signal, signal light, robot or semaphore’ is a signalling device which is positioned at road intersections, pedestrian crossings and other locations to control competing flows of traffic. Traffic lights alternate the right of way of road users by displaying lights of standard colours (red, amber and green) using a universal colour code.
Transportation: the action of transporting someone or something or the process of being transported.
