Time-Variant Travel Time Prediction Model and Its Application in a Cooperative Traffic Control System

Abstract

This paper presents a travel time prediction model and an application of this model to vehicle cooperative control systems. The travel time prediction model considers vehicle delay because of downstream queue formation and traffic signal operation. For estimation of traffic signal–induced delay, this model uses a time-variant nonlinear probability function that maps the vehicle arrival time at a downstream stop line to the probability that it will receive a green signal on arrival. The probability function is determined iteratively. The travel time prediction model can be applied to any phase-based traffic controller. The road vehicle cooperative control system is developed on the basis of vehicle-to-infrastructure communication. The control system optimizes signal timing according to data transmitted in real time and sends speed adjustment recommendations to selected vehicles on the road. Results from numerical tests validate the accuracy of the travel time prediction model and demonstrate the benefits of the cooperative control system.