A method of vehicle trajectory clustering based on graph theory

Abstract

The invention relates to a vehicle trajectory clustering method based on graph theory. Firstly, the vehicle networking data is obtained, and the massive vehicle trajectory data is cleaned by using a Spark platform. Secondly, the coordinate points are projected onto the map, and the connected graph is formed according to the set K value and the relative approximation through the relative distance relation of the coordinate points, and then the clusters whose connectivity strength is greater than the preset value are merged by using the interconnectivity; finally, the real taxi trajectory data are analyzed to get the traffic flow of different time periods and different areas, that is, the optimal taxi-hailing scheme is given. The invention can effectively improve the speed and quality of data processing.

Description

technical field [0001] The invention relates to the field of vehicle networking and computing, in particular to a graph theory-based vehicle trajectory clustering method. Background technique [0002] With the rapid expansion of the scale of the application system in the vehicle industry, the data generated by the vehicle application is growing explosively, and the intelligent transportation system is facing huge challenges in data collection, processing, analysis, and utilization. In the context of the rapid development of information technology and cloud computing, the concept of Internet of Vehicles has gradually become familiar to people. As a typical application of the Internet of Things, the Internet of Vehicles makes it possible to finely collect the spatio-temporal information of vehicles running on the road network, thereby realizing more intelligent management of traffic. [0003] Cluster analysis method in data analysis technology is an effective method to study ...

AI Technical Summary

Problems solved by technology

[0004] Traditional clustering methods such as DBscan and GMM algorithms have certain difficulties in obtaining clustering conclusions when the sample size is large

Since the relative coefficient is based on the internal structure of the data set to establish an index that reflects the internal relationship between the data, in practice, although there is a close relationship between them in the obtained data, there is no internal relationship between things. At this time, it is obviously inappropriate to obtain the results of cluster analysis based on the similarity coefficient, but the traditional cluster analysis model itself cannot identify such errors

Images