Traffic Length Data Based Signal Timing Calculation for Road Traffic Signals Employing Proportionality Machine Learning

Abstract

A machine learning automaton is created to give varying signal durations for Red and Green lights for different times of the day based on the traffic intensity variation. The duration of Red and Green signals is being constantly varied throughout the day based on the traffic intensity which varies all through the day. A set of camera based or ultrasonic sensors are used to first map the traffic intensity to time duration of Red and Green signals. One economical advantage of this system is that the sensing system can be deployed for a few days, the information collected to create a training data set and once the automaton is trained, it can be deployed.