REAL TIME ROAD TRAFFIC PREDICTION

Abstract

Transportation engineering has evolved into a broad multidisciplinary field during the last few decades. This multidisciplinary nature of the profession has become more profound and visible since the advent of the Intelligent Transportation Systems (ITS) early last decade. The remarkable advances in computers, communications, electronics, control, and other related technologies have found important applications in the transportation system. This thesis models one of the most important aspect of ITS i.e. prediction of the road traffic. In this thesis classification of the road traffic is done using Gaussian Mixture Model-Expectation Maximization (GMM-EM) and Prediction of road traffic is done using Hidden Markov model (HMM) . The proposed approach is neither based on tracking nor on vehicle detection hence, this approach has given new heights to the traffic monitoring systems. Apart from this, this method does not assume or draws analogies of traffic moving as particles, neither does it impose restriction on road conditions or road tributaries and distributaries. The results so obtained are verified and validated using the traffic simulator software.