A Road Map for Geotechnical Monitoring of Transportation Infrastructure Assets using Three-Dimensional Models Developed from Unmanned Aerial Data

Surya Sarat Chandra Congress,Anand J. Puppala

DOI: https://doi.org/10.1007/s40098-020-00470-y

2021-02-01

Indian Geotechnical Journal

Abstract:Infrastructure forms the backbone of a nation's growth and economy. The smooth operation of those infrastructure assets depends on many factors including proper use of construction materials under different loading, environmental and durability assessments, and the frequency of monitoring after construction. The performance and maintenance of an infrastructure asset thus depend on the behavior of the system in a built geological environment. Proactive monitoring of infrastructure often leads to preventive maintenance. However, it is not economically feasible to use the current traditional monitoring techniques, especially considering the vastness of the infrastructure networks. In this study, unmanned aerial vehicle–close-range photogrammetry (UAV–CRP) technology is being proposed as a supplemental data collection tool to complement existing traditional monitoring techniques for geotechnical infrastructure. Two case studies covering a pavement structure built over rehabilitated subgrade rich with sulfates and stability of a rock slope adjacent to an old rail line were monitored aerially to understand their state of health conditions. The pavement site had a history of experiencing sulfate-induced heaving and was rehabilitated using extended mellowing after lime stabilization. The rail line under inspection was constructed more than a century ago, and the stability of the weathered rock cut holds the key for safe operations on the rail line. The rock was highly weathered and considered to undergo circular failure. Aerial images were collected and processed to build three-dimensional models to evaluate and assess the condition of these geotechnical assets. This approach not only provides a comprehensive idea through dense point cloud models offering real-field-like view of the asset conditions but also is expected to result in significant savings in data collection time and costs.