Bilkent University
Mechanical Engineering Department
Unraveling The Mechanics of Architected Soft Systems
ABSTRACT
This talk explores the control and dynamic modeling of soft, slender bodies and their interactions with fluid flow. Specifically, it
focuses on developing scalable software and algorithms to model two-way flow-structure interactions between soft, slender structures
with viscous flows. Leveraging these tools alongside topological and geometric principles, this work aims to unravel how the complex
muscular architecture of octopus arms simplifies control, with potential applications in designing dexterous soft robots. This is
achieved by a scalable, user-friendly, open-source software platform called PyElastica. This platform models the dynamics of soft,
slender structures using Cosserat rod theory, which efficiently captures the 3D dynamics of slender objects through a one-dimensional
Lagrangian framework. Building on this, we introduce algorithms that integrate Cosserat rod theory with the velocity-vorticity
formulation of the Navier-Stokes equations, allowing us to model soft, slender structures immersed in viscous flows. The result is a
validated, multiphysics framework that accurately simulates the behavior of both rigid and soft, homogeneous and heterogeneous
structures in fluid environments, setting the stage for scientific discovery and engineering design. Next, we apply these frameworks to
gain insights into the exceptional dexterity and reconfigurability of muscular hydrostats, such as octopus arms, through numerical
simulations. To achieve this, we develop a highly detailed model of an octopus arm with over 200 continuous muscles, informed by
imaging experiments. We demonstrate that complex 3D arm motions are driven by the storage, transport, and transformation of
topological and geometric properties, achieved through simple muscle activation patterns. Building on this understanding of the arm’s
mechanics and topology, we automate control by designing a 3D feedback controller. The controller’s performance is demonstrated
through various reaching and tracking benchmarks, including a challenging case where an eight-armed octopus with 1,600 continuous
muscle groups is controlled. Overall, this talk presents novel physical discoveries and versatile numerical algorithms packaged into
scalable software. These contributions enable the accurate modeling of complex fiber-based structures in fluid environments, with
wide-ranging applications in engineering, healthcare, and medicine.
ABOUT THE SPEAKER
Arman Tekinalp is a Postdoctoral Researcher at the Department of Mechanical Engineering at
the University of Maryland College Park working with Eleonora Tubaldi. He earned his Ph.D.
in Mechanical Engineering (2024) under the supervision of Mattia Gazzola, and his M.S. in
Aerospace Engineering (2019) with Deborah Levin, both from the University of Illinois UrbanaChampaign. He received his B.S. in Aerospace Engineering (2016) from Middle East Technical
University. He is a Fulbright Scholarship recipient. His research focuses on understanding
the mechanics of soft, slender structures and developing numerical methods to capture their
dynamics, whether interacting with frictional surfaces or immersed in fluid flows with potential
applications in designing dexterous soft robots. His work has been published in journals such
as PNAS and Computer Methods in Applied Mechanics and Engineering, with one of his
publications selected as the cover article of PNAS.
ZOOM DETAILS
https://zoom.us/j/2837443344?pwd=NnZJaEpwQklJdUlxNGZtcFhRY0Rjdz09&omn=99976476223
Meeting ID: 283 744 3344.
Passcode:2354290
Date: May 2, 2025, FRIDAY. Time: 14:00.
CONTACT
Dr. Orcun Koray Celebi, Bilkent University
Email: orcun.celebi@bilkent.edu.tr
View seminar poster for more information: