Discrete time point processes in urban traffic queue estimation

The formation and dispersion of queues at a traffic signal is modeled by a discrete time, time-varying Markov chain which is related to the observation point process from vehicle detectors. Three such models of increasing complexity are given. Recent results in the theory of point process filtering and prediction then give the nonlinear minimum error variance filters/predictors corresponding to these models. It is then shown that these optinmal estimators are computationally feasible in a microprocessor. All three algorithms were tested against the UTCS-1 traffic simulator and, in one case, against an algorithm in current use called ASCOT. Some results of these tests are shown. They indicate good performance in every case and better performance than ASCOT in the comparable case.