Automated measurement of highway retaining wall displacements using terrestrial laser scanners

• Highway retaining wall displacements were measured using laser-scan data.
• Horizontal joints between MSE wall panels were chosen as displacement benchmarks.
• A real-life MSE wall's horizontal joints were extracted with 94.26% accuracy.
• The method was tested for 540 different simulation scenarios.
• The wall displacements were measured with an average error of 0.9 mm.

Highway retaining walls are continuously monitored during their construction to ensure the required performance criteria are met. Currently, inspection data are obtained manually by using geotechnical field instrumentations and surveying equipment. The data collection practice is often time-consuming, and erroneous. High precision and accuracy of laser scanners have made them valuable in construction, especially for structural health monitoring and change detection. This paper proposes a new method to (1) extract geometric features of highway retaining walls from the laser-scan data; and (2) use the extracted features as benchmarks for detecting wall displacements. The method is evaluated using Mechanically Stabilized Earth (MSE) walls, which are one of the most widely used highway retaining walls in the United States. Specifically, we investigated the hypothesis that the horizontal joints between the panels of an MSE wall could be used in point clouds to detect the wall's displacement with the required accuracy for condition assessment purposes. First, the proposed methodology was tested using the actual point cloud data of an MSE wall. The proposed feature extraction method was able to extract the horizontal joints with 94.26% accuracy. Then to test the generalizability of the method, we created 3D models of an MSE wall by using the wall's actual dimensions and geometric parameters. These models were used to examine the validity of the hypothesis for 36 laser scanner settings and 15 wall displacement scenarios. Laser scanner sensor parameters along with the 3D models were imported into a simulation environment to generate synthetic data to evaluate accuracy of the proposed method. The method was also tested with 3D models of two more types of MSE walls for a selected number of simulation scenarios. The results demonstrated that the proposed method is capable of measuring wall displacements with an average error of 0.9 mm.