Estimation of Real-time Traffic State of City Expressway under Incomplete Information

According to the characteristics of city expressway traffic flows,the macroscopic traffic flow models of the basic stretch,on-ramps and off-ramps stretch,weaving section in the city expressways are presented respectively in this paper,in which the traffic measurement model is also be proposed,and these models are organized in a state-space form,based on which a traffic state estimator is designed by use of the Extended-Kalman-Filter method.The model parameters are converted into state variables during the estimation process,so the model parameter can be dynamically tracked and the estimation accuracy is improved.The experimental results show that,even by using a limited amount of measurement data,the traffic state of the whole stretch can be obtained satisfactorily.