Road traffic state acquisition method based on Kernel-KNN matching

Abstract

The invention discloses a road traffic state acquisition method based on Kernel-KNN matching. The road traffic state acquisition method based on the Kernel-KNN matching includes that firstly, extracting representative road traffic data, and building a road traffic running character reference sequence after processing the data; secondly, selecting a road traffic data sequence, and building an Mercer core of the road traffic data sequence; finally, selecting a road traffic character reference data sequence and a current road traffic character sequence, acquiring the Euclidean distance between the road traffic character reference data sequence and the current road traffic character sequence in a character space, selecting k road traffic states through a KNN method, and acquiring the road traffic state through weighted average. The road traffic state acquisition method based on the Kernel-KNN matching realizes the real-time and effective dynamic estimation for the road traffic running state. The road traffic state acquisition method based on the Kernel-KNN matching is capable of effectively acquiring the road traffic state and good in precision.

Description

technical field [0001] The invention belongs to the field of traffic state acquisition, and relates to a road traffic state acquisition method. Background technique [0002] Acquisition of road traffic status is an important prerequisite for traffic management such as traffic flow control and guidance. It is the necessary basis for formulating traffic safety management strategies, traffic accident detection, and traffic accident cause analysis. And maintenance indispensable first-hand information. Therefore, traffic status acquisition is a fundamental and important issue in traffic management, traffic safety assurance, and traffic infrastructure monitoring and maintenance. [0003] In the existing road traffic state acquisition methods, due to the complexity and variability of the road traffic system, methods such as parameter models and artificial intelligence often have many restrictive conditions, and it is difficult to achieve extensive road traffic state acquisition. ...

AI Technical Summary

Problems solved by technology

[0004] In order to overcome the shortcomings of the poor accuracy of existing road traffic state acquisition methods, the present invention provides a road traffic state acquisition method based on Kernel-KNN matching that effectively obtains road traffic state and has better accuracy.

Images