Estimation Of Dynamic Origin-Destination Matrix Based On Forcasting Information

Intelligent Transportation Systems (ITS) is the indication that transportation has entered the information era, which has been paid much attention in China and other developed countries. Dynamic Origin-Destination (OD) matrix estimation is one of important part in ITS.Dynamic OD matrix describes the time-dependent traffic demand in traffic network. It is the basis of network state estimation and effective traffic guidance. Generally speaking, it is very difficult to obtain the dynamic OD matrix through traffic investigations which can not predict future network traffic condition and provide guidance information real-time. But the guidance information based on OD estimations influence network state. These influences correspondingly exist in the estimating process for future OD matrix (future OD estimation should be consistent with future network state). A theoretical framework for real-time traffic guidance systems is set up, in which the dynamic OD estimation and prediction module and the interrelations between OD estimations and guidance information is an important component. Kalman Filtering algorithm is introduced for real-time dynamic OD estimation, and a consistent anticipatory route guidance framework which is based on fix point theory and method of successive is established in this thesis.In order to provide a practical test bed for dynamic OD estimation and anticipatory route guidance, the model of Dynamic Network Analysis (DNA) combined with the theoretical framework for real-time traffic guidance system is represented. A new dynamic shortest path (subpath) algorithm is established in DNA; a new meso "TrafficStream" model is established to better simulate network flow in DNA; a new traveler route choice model combined with guidance information is established in DNA. DNA is implemented with C++ based on above models and algorithms. DNA is a meso traffic simulator for small area traffic network and tested on a test network in which is 5 links and 6 nodes for dynamic OD estimation algorithm.