Traffic signal control method based on SVM and computer network and control system

Abstract

The invention provides a traffic signal control method based on an SVM and a computer network and a control system. Firstly, the total vehicle flow and the vehicle flows of straight driving, leftwardturning and rightward turning of all lanes which are collected by a camera and an electromagnetic sensor are respectively utilized as data sets, and the SVM is utilized for separately training 95% ofeach data set to obtain SVM models of straight driving, leftward turning and rightward turning; secondly, 5% of each data set is utilized for calculating a fitting error of each SVM model, according to the fitting errors, a training set is expanded, and the SVM models are updated; finally, the total vehicle flow, collected by the camera in real time, of a crossing is input into the SVM models of various directions, vehicle flow prediction values of straight driving, leftward turning and rightward turning at the crossing are output, according to the vehicle flow prediction values, the passing time of all the lanes is distributed and controlled through the computer network, and regulation and control over traffic signals are achieved. According to the traffic signal control method, the passing time is distributed and controlled according to the vehicle flow prediction values of all the lanes, area planning is achieved, the peak-hour road congestion phenomenon is effectively reduced, andthe passing continuity is improved.

Description

Technical field [0001] The invention relates to the technical field of traffic signal control of a computer network, in particular to a traffic signal control method and control system based on SVM and computer network. Background technique [0002] Support vector machine SVM is a machine learning method based on statistical learning theory. It can analyze data, identify models, and be used for classification and regression. Support vector machine SVM requires few training samples and has strong generalization ability. Support vector regression uses a kernel function to nonlinearly transform the input space into a high-dimensional feature space, and then find the optimal classification plane in this high-dimensional feature space to minimize the distance between each point and the regression line, and then obtain the functional relationship between input and output. Computer network refers to the connection of multiple computers with independent functions in different geographica...

AI Technical Summary

Problems solved by technology

[0004] Aiming at the problem of traffic congestion during peak hours and single control method of traffic lights, resulting in poor continuity of vehicle traffic, the present invention proposes a traffic signal control method and control system based on SVM and computer network, which can autonomously allocate and control traffic according to the traffic flow of different lanes The passing time of each lane realizes regional planning, effectively reduces road congestion during peak hours, and improves the continuity of traffic

Images