Using Neural Network to Identify Pavement Structure Based on Radar Output

by Busby Attoh-Okine, Univ of Kansas, Lawrence, United States,



Document Type: Proceeding Paper

Part of: Pacific Rim TransTech Conference—Volume II: International Ties, Management Systems, Propulsion Technology, Strategic Highway Research Program

Abstract: Knowledge of asphaltic layer thickness and structure are important in many areas of pavement management. Accurate thickness and structure measurements are needed throughout the roadway network to improve pavement performance prediction, to establish structural load capacities and to develop maintenance and rehabilitation priorities. Ground Penetrating Radar is a non-contact technique that has the potential to survey pavement thickness and structure, while operating at highway speed. GPR is one method that could greatly facilitate subsection identification of pavements. This paper proposes the use of an Artificial Neural Network (ANN) for interpreting radar thickness profile output from pavement thickness and structure surveys. A key contribution of the proposed ANN is the ability to identify the subsurface structure of the pavement without any destructive coring.

Subject Headings: Thickness | Radar | Asphalt pavements | Neural networks | Pavement overlays | Load bearing capacity | Geomatic surveys

Services: Buy this book/Buy this article

 

Return to search