【摘 要】
：
Pavement friction and texture characteristics are important to road surface safety.Despite extensive studies conducted in the past decades,the relationship between pavement texture and surface frictio
【机 构】
：
Southeast University
论文部分内容阅读
Pavement friction and texture characteristics are important to road surface safety.Despite extensive studies conducted in the past decades,the relationship between pavement texture and surface friction has not been fully understood.This paper implements deep learning(DL)techniques to investigate the application of pavement texture data for pavement skid resistance and safety analysis.High speed texture profiles and Grip Tester friction data are collected simultaneously on High Friction Surface Treatment(HFST)sites with various types of lead-in and lead-out pavement sections distributed in 12 states.FrictionNet,a Convolutional Neural Network(CNN)based DL architecture,is developed to predict pavement friction levels directly using texture profiles.This architecture is composed of six artificial neuron layers including two convolution layers,three fully-connected layers,and one output layer,with 606,409 tuned hyper-parameters.50,400 pairs of texture and friction data sets are gathered for training,while another 12,600 pairs for validation and testing.The input of FrictionNet is the spectrogram of original texture profile for 1-meter segment,and the output is the corresponding friction level ranging from 0.1 to 1.0.The FrictionNet achieves 96.85%accuracy for training,88.92%for validation,and 88.37%for testing in friction prediction.The result demonstrates the potential of using DL methods for highway speed non-contact texture measurements for pavement friction evaluation.
其他文献
随着近年来的急速发展，城市空气污染问题越发突显。要明确的掌握污染的来源并做出决策，明确各类源项对污染问题贡献是重要手段。但大气污染源种类繁复，要对所有源项进行细致排查难以实现。随着GIS 和地图产品的丰富，地理信息大数据为更便捷详尽的掌握地面源项信息提供了可能。本文以成都为例，对地理信息数据进行分类并建立了其同各类源项的表征关系，基于缓冲区分析和回归分析，分析了各类源项同臭氧浓度的相关性，明确了各
Traffic forecasting is increasingly playing an important role in traffic management.However,current prediction models are typically performed for individual road links and prediction horizons.Differen
现代城市综合交通系统每天产生海量庞杂、异质多源、大范围时空关联的数据，蕴含着具有丰富价值的信息。我国智能交通系统建设一直以来存在交通基础数据库与交通解决方案严重脱节的问题，从而制约了城市综合交通系统的整体效能。通过构建基于大数据的城市多模式交通系统协同优化平台，能够有效整合跨部门、跨区域的交通大数据资源，并充分挖掘海量数据中蕴含的价值信息，突破综合交通大数据应用关键技术，对未来城市多模式交通系统的
Many governments manage a large pavement network regarding infrastructurecondition assessment and maintenance planning.Pavement Management Information System(PMIS)is used to conduct efficient pavement
Bus bunching refers to the phenomenon that several buses arrive at a station within a short time period.Bus bunching often occurs on urban bus routes.It greatly increases passengers waiting time and r
Bicycling is a potential way to improve sustainable urban transportation system.However,the evaluation methods and metrics for bicycle ride quality and slow traffic pavement condition are still in the
共享单车作为一种新兴的城市交通方式，以其便捷环保等特点，正在改变着市民的出行习惯，支持和实现了交通结构上的突破，也给城市规划者和管理者发展可持续城市交通引导了一个新方向。共享单车逐步在全国大部分城市得到大范围布设和使用，很好地解决了市民“最后一公里”的问题，但优势和问题并存，随着共享单车在国内大部分城市的推广和发展，其带来的使用和停放等问题也给很多城市带来了不小的挑战，引起了社会的广泛关注和热议。
With the increase of carbon emissions from urban transport,the topic of carbon emissions cannot be ignored.A bi-level programming traffic assignment model in consideration of carbon emissions charging
Combined traffic assignment-signal control equilibrium is usually integrated into a noncooperative games between the network authority and road users.Unlike a pure Wardropian equilibrium,there might b
将马尔科夫链应用到机场道面外观质量预测中,为机场道面维修决策服务,适用于解决机场道面的历史资料匮乏,难以进行准确可靠预测的问题。在划分道面外观质量等级的基础上,应用马尔科夫理论,建立道面外观质量的状态转移概率矩阵,并采用统计分析方法确定了其计算过程,将灰色GM(1,1)预测模型与道面外观质量状态转移矩阵结合使用,最终建立马尔可夫链预测模型。通过实例的计算分析,确定在8 年时道面外观质量恶化程度严重