Towards Online Quasi-dynamic o-d Flow Estimation/Updating

The paper deals with the proposition of a Kalman filter specification for quasi-dynamic estimation/updating of o-d flows from traffic counts, i.e. under the assumption that o-d shares are constant across a reference period (i.e. a quasi-dynamic interval), whilst total flows leaving each origin vary for each sub-period within the reference period. Drawing upon the effectiveness and the reliability of the assumption of quasi-dynamic o-d flow pattern and of the performances of the quasi-dynamic estimator in offline contexts, the paper illustrates a first formulation of a non-linear quasi-dynamic Kalman filter, which can embed diverse specifications of the state variables and of the corresponding transition and measurement equations. Results of preliminary tests on a synthetic network are presented, and the overall research pattern is also outlined, together with concerned research and practical perspectives.