Scholarlink

<p>Efficient monitoring of the condition of road infrastructure is essential for the provision of a reliable and sustainable mobility and transportation network. The assessment of the structural condition of the asphalt base layer is particularly important in this respect. This paper presents a data-driven approach for an innovative road monitoring system for the non-destructive and continuous determination of the degree of degradation of asphalt roads. The innovation of the project lies in the application of Artificial Intelligence methods to derive the degradation state of the asphalt base layer on the basis of sensor measurements obtained by means of a novel hybrid sensor fabric integrated directly into the asphalt base layer. The proposed Machine Learning-based diagnosis relies heavily on the quality of sensor data. Therefore, we introduce a new method to evaluate the significance of sensor measurements using time series analysis techniques. The feasibility and functionality of the approach is demonstrated through extensive experiments by embedding the sensor material in real asphalt specimens, which are subject to controlled load tests.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Learning road degradation with asphalt-integrated sensor fabric

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年12月5日

查看全文
  • Ralf Bruns,
  • Jürgen Dunkel,
  • Maximilian Greve,
  • Christina Haxter,
  • Joris Herrmann,
  • Sascha Kayser

摘要

Efficient monitoring of the condition of road infrastructure is essential for the provision of a reliable and sustainable mobility and transportation network. The assessment of the structural condition of the asphalt base layer is particularly important in this respect. This paper presents a data-driven approach for an innovative road monitoring system for the non-destructive and continuous determination of the degree of degradation of asphalt roads. The innovation of the project lies in the application of Artificial Intelligence methods to derive the degradation state of the asphalt base layer on the basis of sensor measurements obtained by means of a novel hybrid sensor fabric integrated directly into the asphalt base layer. The proposed Machine Learning-based diagnosis relies heavily on the quality of sensor data. Therefore, we introduce a new method to evaluate the significance of sensor measurements using time series analysis techniques. The feasibility and functionality of the approach is demonstrated through extensive experiments by embedding the sensor material in real asphalt specimens, which are subject to controlled load tests.