Research on Application of Smart Phone and Vehicle Positioning Analysis Model Based on SNN Density ST-OPTICS Algorithm in HOV Lanes

摘要

Due to the long-term infrastructure construction of traditional Chinese HOV lane detection methods and the large investment, the monitoring mainly concentrates in the scope of equipment deployment. Nevertheless, there are still many monitoring blind spots. Based on the ST-OPTICS algorithm of SNN density, a model of mobile smart phone and vehicle positioning analysis and detection was established, which could effectively estimate the number of actual passengers on board a vehicle to a certain extent and provide a new type of vehicle real-time monitoring in HOV lanes. Through the collection of running vehicles on mobile roads and GPS and base station positioning data of mobile smart phones, the original data were preprocessed to obtain a feasible data set, then maps were matched according to their latitude and longitude and other information, and the clustering algorithm was used to classify and establish a positioning analysis model. The passenger metrics of each cluster was calculated, and the data set meeting the conditions based on the HOV was employed as the basis for vehicle and mobile positioning analysis model. Experiments showed that the application of the model in the HOV lane could effectively assist the detection of HOV lane vehicles and the better detection accuracy.