Research On Calculation Method Of Traffic Intersection Similarity Based On Time And Space Characteristics

In recent years,with the construction of smart cities and intelligent transportation,especially the wide application of technologies such as big data,Internet of Things and cloud computing,the infrastructure of urban traffic intersections has become more sophisticated.From the massive historical space-time data of traffic intersections,quickly retrieving the expected data similar to the current intersection status has become an urgent need for research on spatio-temporal data mining,signal timing scheme recommendation and short-term traffic flow prediction.Aiming at the lack of spatial data analysis,low computational efficiency and poor accuracy,the existing traffic intersection similarity calculation method starts from the actual traffic demand,and proposes a traffic intersection similarity calculation method based on time and space characteristics.The spatial similarity,ontology attribute similarity and dynamic data similarity of intersections have been studied in depth.The main work is as follows:(1)The similarity calculation method for traffic intersection space scene is proposed.First,the traffic intersection is placed in a space scene,and the geometric types and spatial relationships within the spatial scene are analyzed.Then,based on the topological relationship,the azimuth relationship and the distance relationship between the spatial geometries,a unified set arrangement standard is used to construct the spatial relationship feature description model.Finally,combined with the research ideas of spatial analysis,the spatial similarity of traffic intersections is calculated.(2)The similarity calculation method of the traffic intersection ontology attribute is proposed.First,according to the characteristics of the traffic intersection,it is determined that the traffic intersection ontology attribute contains the number of state values,and the attribute type is divided.Then define different attribute feature description values for the characteristics of the numeric attribute and the nominal attribute.Finally,the eigenvalues are integrated by the comprehensive evaluationmethod,and the similarity calculation model is constructed to realize the method of calculating the similarity degree of the traffic intersection ontology with high accuracy and practicability.(3)The similarity calculation method for traffic intersection dynamic data is proposed.Firstly,the dynamic data influencing factors between adjacent lanes of traffic intersections are analyzed,and the dependence of dynamic data and intersection ontology attributes is created in the time series database.Then,the intersection body attribute is used as a medium to correlate the dynamic data and spatial data of the traffic intersection.Finally,considering the temporal characteristics and storage problems of spatio-temporal data,combined with the idea of hierarchical clustering analysis and quantitative memory,the dynamic data similarity calculation method with high efficiency,clear hierarchy and adaptability is realized.