摘要
道路车辆拥堵问题导致交通事故增加,降低了居民的出行效率,长时间的道路拥堵更是加重了环境污染,造成国家经济损失等诸多问题。为缓解城市道路交通的拥堵问题,提高出行效率,基于隐马尔可夫模型,针对已有道路拥堵时间数据进行采集与建模,并对该隐马尔可夫模型进行训练,通过算法计算与分析,预测未来一段时间的道路拥堵情况,为人们的出行提供拥堵时间预测,而后提出不同时段通过道路用时最短的最优路径。对韦尔奇算法进行改进,在原算法基础上增加考虑前n时刻状态。利用改进型韦尔奇算法,使得训练集参数更精确,达到预测精度更高的目的。实验结果表明,预测数据结果与真实数据相比,误差不超过3%,该模型预测结果具有较高准确性。
The problem of road vehicle congestion leads to the increase of traffic accidents and reduces the travel efficiency of residents.Long-term road congestion aggravates environmental pollution and causes many problems such as national economic losses.In order to alleviate the congestion problem of urban road traffic and improve travel efficiency,based on the hidden Markov model,this paper collects and models the existing road congestion time data,trains the hidden Markov model,predicts the road congestion time in the future through algorithm calculation and analysis,and provides congestion time prediction for people’s travel,Then,the shortest optimal path through the road in different periods is proposed.The innovation of this paper is to improve Welch algorithm and add the state of the first n moments on the original basis.The improved Welch algorithm is used to make the parameters of the training set more accurate and achieve the purpose of higher prediction accuracy.The experimental results show that the error between the predicted data and the real data is no more than 3%,and the prediction result of the model has high accuracy.
作者
张绪冰
谢雨飞
ZHANG Xubing;XIE Yufei(School of Electrical and Information Engineering,Beijing University of Civil Engineering and Architecture,Beijing 100044,China)
出处
《计算机工程与应用》
CSCD
北大核心
2022年第16期312-318,共7页
Computer Engineering and Applications
基金
国家自然科学基金(61703028)
北京市属高校基本科研业务费(X18071,X19020)
北京市教委科研项目-科技计划一般项目(KM202110016007)。
