Smartphone-Based Pavement Roughness Estimation Using Deep Learning with Entity Embedding

Smartphone-Based Pavement Roughness Estimation Using Deep Learning with Entity Embedding
复制标题

使用深度学习和实体嵌入进行基于智能手机的路面粗糙度估计

DOI:
10.1142/s2424922x20500072
复制
发表时间:
2020
期刊:
Adv. Data Sci. Adapt. Anal.
影响因子:
--
通讯作者:
Y. Adu
Y. Adu
中科院分区:
--
文献类型:
--
作者:
Armstrong Aboah;Y. Adu

文献摘要

被引文献

相似文献

测量路面平整度的常用指数是国际平整度指数(IRI)。传统的路面信息采集方法成本高昂,因此多年来研究人员一直采用其他更便宜的方法来采集数据。本研究的重点是开发一种深度学习模型,以更低的成本快速准确地确定路段的IRI值。该研究提出了一个模型,使用加速度计数据和前一年的IRI值来预测今年的IRI值。该研究的结论是,在前一年的IRI中增加加速度计读数增加了预测的准确性。
The commonly used index for measuring pavement roughness is the International Roughness index (IRI). Traditional method for collecting road surface information is expensive and as such researchers over the years have resorted to other cheaper ways of collecting data. This study focuses on developing a deep learning model to quickly and accurately determine the IRI values of road sections at a cheaper cost. The study proposed a model that uses accelerometer data and previous year’s IRI values to predict current year IRI values. The study concludes that addition of accelerometer readings to previous year’s IRIs increased the accuracy of prediction.