Abstract

Passageways cut through rock might be subjected to rockfalls. If a falling rock reaches the road area, the consequences can be disastrous. The traditional rockfall risk assessment method and risk mitigation are based on on-site investigations performed by a geologist or a rock engineer. The parameters resulting from the investigation, such as discontinuities, orientations and spacings, potential rockfall initiation locations, slope geometry, and ditch profile, are either measured or estimated. We propose a photogrammetry-based method for estimating the probability of failure for rockfall. Several photographs of the rock-cut are taken, and a 3D geometry is computed using photogrammetry. This model already allows remote visual inspection of the site. The information about joint planes can be discovered semiautomatically from the point cloud. Next, the probability of rockfall reaching the road area is computed using probabilistic kinematic analysis on the geometry extracted using photogrammetry. The results can be used to define the rockfall probability for each rock-cut. Furthermore, the results can be used to determine the appropriate rockfall risk mitigation actions for each rock-cut.

Introduction

Falling rocks are a significant risk across all continents [1-4]. Rockfall is a life-threatening hazard for highways, railroads, water passages, and open-pit mines. Rockfall risk assessment is a significant problem because of the vast amounts of assets and the considerable size of each asset. Currently, the risk assessment is carried out empirically based on visual inspections and expert judgement or using simplified analyses with visually obtained data. The subjective process is slow and prone to human errors. Imprecise risk assessment leads to overly conservative designs that are expensive and dangerous, could threaten life and cause economic or ecological problems.

The state-of-the-art technology is to use remote sensing technologies to map rock-cuts to assess rockfall risk [5-16]. The technology and associated software for rapid collection of initial data suitable for rockfall risk assessment already exists and has become a mature and reliable source for initial data for more advanced rockfall analyses [9, 17-21]. Finally, work to codify the automated design processes is already underway [22-24].

This content is only available via PDF.
You can access this article if you purchase or spend a download.