Research And Application Of Urban Traffic Data Acquisition And Fusion

Traffic Guidance System is the kernel part of the Intelligent Transport System. Its success depends on whether or not the multi-source probed traffic data can be fused correctly.There exist many traffic data detecting methods nowadays, but the data exchange is much worse. So the traffic data acquired by those methods are different, and the same parameter may exist different detect results. Furthermore, there lacked appropriate means to analyze these results for obtaining the undersurface information. The traffic information fusion method based on evidence theory is put forward to figure out the problems mentioned above.There are three parts in this paper: 1. A new kind of dynamic OD analysis method is presented here.This method usilizes the sliding window theory to preprocess the snapshot data during the replacement cycle. Finally, the travel time can be obtained as example of the algorithm. 2. The typical route theory is put forward here to reflect the mass travel custom which is important to the traffic administration while lacks effective methods to acquire. An algorithm is researched to obtain the typical route among urban districts. It can haunt the candidate routes by means of heuristic algorithm of hot spots, and then determine the typical route resorting to the fuzzy clustering. 3. A new traffic information fusion method is illustrated. Several groups of original multi-source traffic data are fused into one group of traffic data which has higher reliability by evidence theory. The Greenshields theory, improved attribute reduction and fuzzy probability measure are used to reduce redundancy and obtain the basic probability assignment function. Finally, a new formula is presented to figure out the final step of fusion and the conflict among the evidences.During the research on the Important Project for Key Subject of Zhejiang Province " Research and Application of the Key Issues in Urban Intelligence Traffic Guidance System " (2006C13100) ,these theories and methods above are put into practice. Meantime, the Hangzhou traffic guidance information subsystem is developed which is consisted of three submodels. The practice proved that these theories and methods are available and exact.