Ph.D. Thesis Defense Announcement

BIO-INSPIRED SOIL EXCAVATION AND PENETRATION: IMAGE-GUIDED NUMERICAL MODELING AND EXPERIMENTAL INVESTIGATIONS

By Meron Berhanu Belachew

Advisors:

Dr. J. David Frost (GT – CEE) & Dr. Chloé Arson (Cornell University – EAS)

Committee Members: Dr. Sheng Dai (GT – CEE), Dr. Catherine O'Sullivan (Imperial College London – CEE), Dr. Gioacchino Viggiani (Université Grenoble Alpes – Laboratoire 3SR)

Date and Time: April 20, 2026. 7:00AM EST

Location: SEB122 or https://gatech.zoom.us/j/94160328747

Subsurface excavation and penetration are central to geotechnical engineering
applications including trenchless technologies, tunneling, site characterization,
anchoring, and underground infrastructure installation. Conventional engineering
approaches often rely on tunnels of simple geometries and linear paths, as well as
brute-force pure soil removal or pure soil displacement. These methods often
require high energy input and induce significant ground disturbance. In contrast,
many biological systems achieve efficient underground construction and
locomotion through controlled geometry, staged excavation, and adaptive
subsurface advancement combining excavation and penetration. This thesis
investigates how such natural strategies can inform the analysis and design of bioinspired
geotechnical systems.
The research focuses on three complementary studies. First, the architecture of
Harvester ant nests is obtained using three-dimensional scanning of in-situ castings
to identify geometric characteristics relevant to underground stability. Shaft Georgia Institute of Technology
School of Civil and Environmental Engineering
Atlanta, Georgia 30332-0355 U.S.A.
Phone: 404.894.9044
A Unit of the University System of Georgia • An Equal Education and Employment Opportunity Institution
geometries, chamber shapes, sizes, and spatial arrangement are examined, and
their influence on stress redistribution in the surrounding soil is evaluated through
numerical modeling and analytical solutions. The analyses show that ant nest
structures can be reasonably represented using mathematical models that mimic
the natural systems, and that their geometry and spatial distribution enhance the
mechanical stability of the surrounding soil.
Second, the thesis investigates staged excavation strategies inspired by ant nest
construction and compares them with conventional single-pass excavation. Finite
element analysis is used to study the effects of excavation sequencing on stress
plastic strain development, and energy demand. The results demonstrate that
staged excavation can reduce the mechanical work required relative to single-pass
excavation under comparable conditions, although with slightly higher and more
localized plastic shear strain.
Third, a bio-inspired soil penetration concept motivated by root growth, centipedelike
motion, and vortex geometries is developed and studied. This compound
concept integrates functions observed in biological systems into a solution that lies
between pure excavation and pure penetration, with the goal of reducing energy
demand, penetration resistance, and subsurface disturbance. Experimental devices
are designed and tested in granular soils, including systems compatible with X-ray
computed tomography imaging. Time-lapse photography, digital image correlation,
X-ray CT, continuum-based and particle-scale digital volume correlation are used to
characterize soil deformation at the micro-scale and interpret it alongside macroscale
measurements of external response measurements and work input. The
results show that the friction-reversal component of the device can generate selfpenetration,
while combined linear and rotational motion with vortical tip geometries
can reduce disturbance, energy demand, and tip penetration resistance, while also
expanding the information that may be retrieved from in-situ penetration tests.
Overall, this thesis demonstrates that biologically inspired excavation and
penetration strategies provide a useful framework for rethinking geotechnical
systems. By combining numerical modeling, experimental mechanics, and imagebased
characterization, the work establishes links between natural subsurface
construction and engineering implementation. The findings contribute to both the
fundamental understanding of soil-structure interaction and the development of
more efficient, lower-disturbance geotechnical technologies.