An Automatic Traffic Incident Of Freeway Detection Algorithm Based On Dynamic Traffic Model And Multi-information Fusion

With the development of highway and traffic problem being serious day by day in big cities,the demand of establishing Emergence Management Systems (EMS) becomes urgent.Effective detection for traffic incident is the first step and the key component of operating the EMS successfully;therefore,the study on detection algorithm for traffic incident has become a hot issue.The study tries to combine the theory and neural network,on the basis of dynamic traffic model to establish a automatic detection algorithm for traffic incident using neural network based on wavelet theory filtration.Main research work is listed as follows:The existing automatic detection algorithms for traffic incident are summarized,the performances of it are compared and analyzed,the index for evaluating algorithm's performance is listed and then the basic principle of the algorithm in the thesis is expounded.In data low-grade processing stage,this thesis utilizes wavelet theory for processing the data collceting by loop coil vehicle sencor and compares with three methods effects. Several algorithms using neural network to detect traffic incident also had been introduced and compared.The structure's selection and training of neural network were studied in particular,which were the basis of establishing the algorithm in the thesis.A automatic detection algorithm for traffic incident using ART2 neural network based on wavelet theory filtration was proposed,which has the advantages of neural networkand wavelet theory.Neural network has several characteristics,such as disposing datum in parallel,aiming for overall function,storing information distributed and so on,which can produce a non-linear map by training and learning,cluster data adaptively,with the abilities of restraining the noise's disturbance and good robustness. Wavelet theory can eliminate the noise and redundant targets in sample.The merge not only reduced the scale of the network,lessened burden of training and learning by eliminating redundant targets,but also improved the accuracy of detection by eliminating noise. In order to verify the effect of the algorithm proposed in the thesis,simulation model of traffic incident was established to obtain the traffic data which was used to train and test the algorithm.Comparing and analyzing had been done among the algorithm in the thesis,the existing neural network and traditional detection algorithms.The results show that the algorithm in the thesis has higher detection rate and lower false detection rate.Its coordination ability to evaluation index is also better than other algorithms.