Fuzzy logic for traffic management

How to manage traffic in today's world has become a necessity, so that the traffic lights in cities and on the streets are considered to be the main tools of urban traffic management.
Currently, the method of applying the duration of green and red lights for each route is applied according to the pre-determined rules that can determine the duration of stopping and moving for cars and pedestrians.

Now in the new era and with the advancement of technology and the emergence of a new generation of technology called artificial intelligence, it has given us the opportunity to use these tools in any field we need and use them in various issues.
In this project, we will examine how to use a type of artificial intelligence that uses criteria such as the number of cars that are added to the desired point at the designated time, which is the place where the traffic light is installed, and also the length of the queue of cars that have already arrived have existed at the desired point, they are used to determine the time of traffic lights and based on fuzzy rules.

As we know, fuzzy rules are actually more realistic, so we use these rules a lot in our daily life. The basis of using fuzzy rules is that it examines existing conditions and data by assigning membership values, and this makes conditions to be examined based on probabilities, that is, the same practice that we use in our daily life we are facing it.

In this project, we implement the project using the available library for fuzzy rules and the powerful Python language. At first, we consider the four main factors in determining the time of the traffic light, which are: the balance of both lanes, the imbalance of both lanes, the balance of the first lane, and the imbalance of the second lane, so that both high traffic show for one route, the first route is balanced and the second route is unbalanced so that each shows high traffic for different routes. Being balanced means that the difference in the number of cars for two routes is less than 4 cars, otherwise it will be unbalanced, so we have to examine this balance condition for both the car entry rate and the car queue.




For fuzzy rules, membership values are defined according to the images shown below.




Now, by implementing an artificial intelligence system in such a way that it obtains the values of the arrival rate and the queue length in a certain period of time and gives it to our system, it calculates the duration of the traffic light staying on using the existing membership values and defined calculations the most appropriate time will result.
After defining the membership values, it is time to define the fuzzy rules, based on these rules, the system is supposed to take the input values and determine the time to control the traffic, some of these rules are listed below.




After defining the rules and applying them to the fuzzy system, it is time to give the inputs to the system and see the output. Below is an example of the output of the implemented fuzzy system, which can be seen in the best possible way it determines the time the traffic light stays on for each route, which is the most optimal mode based on the given inputs.




This project shows only one example of the uses of artificial intelligence and fuzzy rules, which can be very useful and used in everyday life. Therefore, in today's world, it can be said that artificial intelligence has the first word and learning how to use this technology is an important and necessary thing in today's era.

Thanks.