zemahran/smart_traffic_management_system

The current traffic management system in most of the Indian cities are either based on the fixed timer control mode or the manual control traffic management. As technology becomes mature Machine Learning is one of the emerging application of Artificial Intelligence nowadays and its infrastructure is also increasing, it can be helpful for solving many traffic problems, in this paper we propose Reinforcement Learning approach with the help of Convolution-Neural-Network to minimize the average waiting time of vehicles at an intersection.

smart_traffic_management_system

The current traffic management system in most of the Indian cities are either based on the fixed timer control mode or the manual control traffic management. As technology becomes mature Machine Learning is one of the emerging application of Artificial Intelligence nowadays and its infrastructure is also increasing, it can be helpful for solving many traffic problems, in this paper we propose Reinforcement Learning approach with the help of Convolution-Neural-Network to minimize the average waiting time of vehicles at an intersection.

Software Tools Used :

SUMO/Python IDE

Virtual Machine Environment on Anaconda ,we needed to install tensorflow to support keras API.

Matplotlib, logfile.txt,Hp5 file to store data in binary form.

To Run this code: step 1: Activate your virtual environment .venv\scripts\activate

step 2: open cmd and Change the current working directory to your local project for example : cd desktop

step 3:RUN your .py file C:\Users\PROJECT\Desktop>python traffic_light_control2.py