微型电脑应用2025，Vol.41Issue(5)：29-32,4.
基于浮动车辆数据的交通流量估计方法
A Traffic Flow Estimation Method Based on Floating Car Data
摘要
Abstract
Floating car data(FCD)provides valuable data for road traffic management.However,the data are incomplete as it only counts vehicles equipped with GPS,model conversion is required.In order to improve the accuracy of traffic flow predic-tion,this paper proposes a method for estimating traffic flow based on FCD.This paper combines FCD with loop detector data,uses a multivariate linear regression approach and determines weights by the random forest algorithm.Based on the traffic data of Qingdao City in 2020,integrating features such as time variables,road function categories and capacity,the regression coef-ficients for each data sample are dynamically adjusted through the soft voting strategy.The research results demonstrate that the preposed method outperforms traditional linear regression and artificial neural network methods,and offers simpler calibra-tion,and usage processes.关键词
交通流量估计/城市交通网络/机器学习/交通模式/浮动车辆数据/线性回归Key words
traffic flow estimation/urban traffic network/machine learning/traffic pattern/floating car data/linear regression分类
信息技术与安全科学引用本文复制引用
张辉..基于浮动车辆数据的交通流量估计方法[J].微型电脑应用,2025,41(5):29-32,4.基金项目
甘肃省知识产权计划(22ZSCQ034) （22ZSCQ034）
