Research on Traffic Flow Fuzzy Time Series Forecasting Algorithm of Single Intersection

Because of the highly nonlinear of the Traffic flow prediction, it is very difficult to build its model. In this paper an algorithm which is based on fuzzy time series forecasting, is used to overcome the difficulty. And we use Coil-based virtual video detectors detect the traffic flow as a input of fuzzy time series prediction, this can make the historical data collection faster and more accurate. The simulation proved it is better than others.