A stochastic estimation approach to real-time prediction of incident effects on freeway traffic congestion

摘要

Real-time prediction of the effects of freeway incidents on traffic congestion is urgently necessary for the development of advanced freeway incident management systems. This paper presents a stochastic esti- mation approach to real-time prediction of time-varying delays and queue lengths which are regarded as two significant variables in examining freeway incident congestion in this study. In addition to system specification utilizing four groups of proposed lane traffic variables, a stochastic estimation approach which involves a discrete-time nonlinear stochastic model and an algorithm based on Kalman filtering is devel- oped to estimate real-time delays and queues in the presence of freeway incidents.