Dynamic Origin-Destination Estimation without Historical Origin-Destination Matrices for Microscopic Simulation Platform in Urban Network

As the input of dynamic traffic assignment (DTA) model, dynamic origin-destination (OD) matrix is crucial for intelligent transportation applications. This paper presents a solution scheme for Dynamic OD estimation without historical OD matrices by using traffic flow data of loop detectors. Through a bi-level optimization structure, OD matrices of different time intervals are estimated sequentially. A ridge regression method is applied to obtain the initial OD matrix in this paper. A constrained nonlinear programming method is proposed to calibrate the assignment matrix for accurately mapping the demand to the traffic flows. With assignment matrix, a modified simultaneous perturbation stochastic approximation (SPSA) algorithm, called Restart-SPSA, is proposed to estimate OD matrix in each time interval. The capability of the proposed solution scheme is validated in an urban network of Singapore using the microscopic traffic simulator VISSIM. The simulation results show that calibration of assignment matrix and Restart-SPSA can improve the performance significantly compared with the method using uncalibrated assignment matrix and original SPSA.