Bulletin of the American Physical Society
APS March Meeting 2024
Monday–Friday, March 4–8, 2024; Minneapolis & Virtual
Session D38: Robophysics IFocus
|
Hide Abstracts |
Sponsoring Units: DBIO DSOFT Chair: Chen Li, Johns Hopkins University Room: 103D |
Monday, March 4, 2024 3:00PM - 3:36PM |
D38.00001: Electromechanical Enhancement of Live Jellyfish for Ocean Exploration Invited Speaker: John O Dabiri Electromechanical Enhancement of Live Jellyfish for Ocean Exploration: |
Monday, March 4, 2024 3:36PM - 3:48PM |
D38.00002: Mechanical characterization of bio-inspired flagella interaction Zhuonan Hao, Sangmin Lim, Siddharth Zalavadia, Darren Chin, Sumukh Johri, Vinay Nagappala, Khalid Jawed Locomotion of bacteria and other microbes has been investigated since their discovery, particularly their movement in low Reynolds flow. Bacteria use flagella-driven mechanisms to swim and turn, such as buckling, bundling, and tumbling. Researchers have used different techniques to investigate these locomotion modes, from numerical modeling to biological experiments and scaled-up setups. While there have been successful measurements of forces on biological flagella, comprehensive analysis of forces resulting from the interaction of multiple elastic flagella remains limited. This research focuses on the role of flagella elasticity, forces involved in fluid-structure interactions, and potential applications in medical microbots. We use the Discrete Elastic Rod (DER) method to model the flagella as Kirchhoff's elastic rods, coupled with the Regularized Stokeslet Segments (RSS) method for the hydrodynamics and the implicit contact model (IMC) for a physical contact simulation. An experimental setup with high precision is developed to measure forces in a viscous medium, and the results are compared to mathematical models to ensure model accuracy. We present a characterization map of propulsive force based on various elastic flagella interactions by varying their number, spacing distance, and motion pattern. |
Monday, March 4, 2024 3:48PM - 4:00PM |
D38.00003: Wave propagation in a basally-actuated robotic filament Rebecca N Poon, Clément Moreau, Benjamin Walker, Daniel Soto, Christopher J Pierce, Eamonn Gaffney, Daniel I Goldman, Kirsty Y Wan Breaking of time-reversal symmetry is necessary for drag-based propulsion at low Reynolds number. Many eukaryotes self-propel using active filaments (particularly cilia), which propagate asymmetric beat patterns though the fluid due to the coordinated action of dynein motors. For a flexible passive filament however, purely basal actuation is generally thought to be insufficient to achieve the large-amplitude bending waves observed in biological cilia. However, in the context of engineering applications, the actuation of filaments via an applied oscillating torque at the proximal end is an effective and parsimonious design. Here we present a simple realisation of enhanced wave propagation in a macroscale robophysical model of a cilium. The artificial cilium, about 5cm in length, beats in a high viscosity fluid to recover low Reynolds number fluid mechanics. Building on previous designs (Diaz et al, 2021), our robotic cilium consists of multiple passive rigid segments connected by hinges and is symmetrically driven by a basal motor. By introducing mechanically asymmetrical hinges, we produce an asymmetrical beat pattern from purely symmetric basal driving. We investigate the effect of link-number on the beat pattern, and compare the resulting waveforms with predictions from a theoretical and computational model of a proximally-driven filament with differential bending stiffness. Finally, we investigate the hydrodynamic pairwise interaction of two cilia. |
Monday, March 4, 2024 4:00PM - 4:12PM |
D38.00004: Octopus arms chasing in fluids Arman Tekinalp, Tixian Wang, Ilia Nasiriziba, Udit Halder, Prashant Mehta, Mattia Gazzola Octopuses employ their arms to detect and hunt preys, through a sophisticated interplay between sensing, actuation and environment. Even when vision is blocked, octopuses can robustly coordinate their boneless arms thanks to an array of chemo-sensors. Here, we combine a 3D computational model of an octopus arm, created from medical imaging and biomechanical data, with sensory feedback control inspired by motion camouflage. The result is a computationally minimal, highly effective strategy, which we demonstrate in a range of tasks, from reaching and grasping to chasing preys’ chemical trails. |
Monday, March 4, 2024 4:12PM - 4:24PM |
D38.00005: Crawling on waves with a soft foot Saravana Prashanth Murali Babu, Ali Sahafi, Jose Bonilla, Ahmad Rafsanjani Snails crawl by generating a sequence of pedal waves via ventral muscle contractions, known as waves, and relaxations termed inter-waves. These actions transfer propulsive forces from a single foot to the ground through a thin layer of mucus secreted by the snail. Inspired by the adhesive locomotion of snails, we have developed a bioinspired soft robot equipped with a single-soft foot. This foot features a network of embedded pneumatic chambers generating various traveling wave patterns, two lateral bending actuators for steering, and distributed fluidic channels for selective artificial mucus deposition. A powerhouse comprised of a compact wireless controller, a pressure sensor, a pump, electro valves, and mucus storage housed in a shell mounted on the soft foot, creating an untethered robot control for robophysical analysis. Here, we show two experimental pathways in analyzing the physics behind snail locomotion. First, we reproduce snail's normal and loping gaits by varying the actuation wavelength and amplitude. Second, through rheological characterizations, we demonstrate that upon actuation, the thin mucus layer between the foot and the substrate creates stress in the layer, leading to adhesive locomotion. Our work integrates soft materials and mucus viscoelasticity to control sliding friction, enabling both direct and retrograde robot locomotion through altering pedal wave direction. |
Monday, March 4, 2024 4:24PM - 4:36PM |
D38.00006: Learning to track flows Haotian Hang, Yusheng Jiao, Sina Heydari, Feng Ling, Eva Kanso Many aquatic animals can follow hydrodynamic trails by sensing and responding to flow signals. Despite numerous studies on this topic, a feedback control strategy that enact this behavior using only local and instantaneous flow sensing remains elusive. Here, we apply deep Reinforcement Learning to solve the problem of following vortical wakes to their generating source. We find that the trained swimmer reaches the source of the wake by two distinct policies: one drives the swimmer toward to region with larger flow speed and the other does the opposite. These policies reveal that flow sensor at tail is crucial for tracking traveling wave signals. Through analysis in a reduced order signal field, we map the sensor location to the stability of the controller in locating the source. Importantly, the sensory control strategy is generalizable to thrust and drag wakes of different Strouhal and Reynolds numbers and to 3D wakes. This work emphasizes the importance of both sensor location and sensor type and has implication on other source seeking control problems with traveling-wave characteristic. |
Monday, March 4, 2024 4:36PM - 4:48PM |
D38.00007: DiSMech: A Simulator for Soft Robots and Flexible Structures based on Discrete Differential Geometry Radha Manoj Lahoti, Andrew Choi, Mohammad Khalid Jawed The study of soft slender structures has garnered substantial interest in diverse scientific realms, including mechanics, robotics, computer graphics, and biomechanics. Owing to their pronounced non-linear deformations under minimal loads, accurately capturing the dynamics of such structures and predicting their motion in real-time presents a formidable challenge. In this research, we present DiSMech, an open-source simulator for intricate soft robots and structures that can be modeled as flexible rods, shells, or interconnected elements thereof. DiSMech leverages Discrete Differential Geometry (DDG) methods, employing discrete elastic rod algorithms for rods and mid-edge normal based shape operators for shells to model the continuum mechanics. A distinctive feature of DiSMech lies in its fully implicit handling of equations of motion, unlike previous simulation frameworks focusing on soft structure physics. This implicit treatment enables significantly larger time step sizes, resulting in simulations that are an order of magnitude faster than existing state-of-the-art counterparts, while preserving the physical accuracy. The framework allows users to design customized geometric configurations using individual elastic rods and shells through an intuitive input interface. For applications in robotics, control is achieved by manipulating the natural curvatures of individual elements. The innovative combination of computational efficiency and physical fidelity positions DiSMech as a powerful tool for studying and advancing soft slender structures in various scientific disciplines and practical domains. |
Monday, March 4, 2024 4:48PM - 5:00PM |
D38.00008: Abstract Withdrawn
|
Monday, March 4, 2024 5:00PM - 5:12PM |
D38.00009: Plant-inspired Mechano-sensing Soft Robot Tofayel Ahammad Ovee, JEAN-FRANÇOIS LOUF, Eftakhar Ahmed Arnob Modern society relies on robotic automation to accomplish menial, hazardous, and sophisticated tasks. A growing subfield of robotics uses soft materials to accomplish these tasks. While promising, soft robots typically lack a key feature: mechanosensitivity. To address this challenge, we designed a smart skin transmitting information using non-linear poroelasticity in a fashion reminiscent of plant mechanoperception. More specifically, we used a biocompatible strain-softening material, Ecoflex©, to design a skin perforated with channels filled with a liquid that we implemented on a homemade 3D-printed arm and connected to a pressure sensor. When the robot arm grasps an object, the skin deformation leads to a superlinear increase in pressure, enabling accurate stress measurements. We could then experimentally access the strain applied by the robotic arm and use these inputs in combination with contact mechanics theory to extract the radius and effective Young's modulus of the grabbed object. Furthermore, the versatile nature of the Ecoflex© sensitive skin enables its implementation on surgical devices to measure the mechanical properties of cancerous and healthy tissues, thus complementing biopsies. As such, this innovative skin broadens horizons for automation, robotic surgery, and ensures safer interactions between robotic devices and humans. |
Monday, March 4, 2024 5:12PM - 5:24PM |
D38.00010: Rubble traversal experiments of a vine robot Zheyu Zhou, Yaqing Wang, Elliot W Hawkes, Chen Li In natural disasters, trapped victims face a 20% survival rate, with near-certain mortality after 48 hours. Current search and rescue robots are poor at going into rubble. Here we developed a low-cost vine robot and studied how well it can grow into simulated rubble. The soft, continuum robot body made from fabric can be steered in 3D with three pneumatic tube actuators along its entire length. The rubble testbed consists of water bottles of variable weights and shapes. We attached LEDs along the robot to observe its growth through the semi-transparent simulated rubble. We tested the robot growing into rubble of varied weights, with its tip not oscillating or oscillating in low and high frequencies. In lightweight rubble, for any tip oscillation treatment, the growing robot could always squeeze through crevices smaller than its tip diameter, and steer effectively. In heavy rubble, without tip oscillation, the robot was unable to fully grow due to limited force capacity; with the tip oscillating in low frequency, the robot could occasionally repel a blocking bottle and slightly enlarge the crevice to negotiate through; with the tip oscillating in high frequency, the robot could not successfully squeeze through due to lack of steady propulsion. |
Monday, March 4, 2024 5:24PM - 5:36PM |
D38.00011: Shape morphing structures with Shrinky Dink and 3D printed Kirigami Mrunmayi Mungekar, M. Ravi Shankar, Mohammad Khalid Jawed We present an equipment-frugal method of creating 3D free-forms from Shrinky Dink sheets, a well-known children's toy, and 3D printed ABS plastic. This involves gluing a piece of 3D printed layer with Kirigami patterns to a Shrinky Dink cutout, forming a flat composite. When this composite is heated, the Shrinky Dink shrinks, but ABS plastic does not. This creates a residual stress between the composite layers, causing it to deform into a three-dimensional shape. Our primary focus is to investigate how the interplay of Kirigami design and the geometric parameters influence the ultimate 3D shape. We present a variety of manufactured shapes inspired by everyday objects, such as a bowl and a spoon. Additionally, we employ established machine learning tools to generate various 3D shapes during the process. By using easily accessible tools like a household oven and a 3D printer for production, our research provides a means to democratize the creation of deployable structures through the accessibility and cost-effectiveness of the manufacturing methods. |
Follow Us |
Engage
Become an APS Member |
My APS
Renew Membership |
Information for |
About APSThe American Physical Society (APS) is a non-profit membership organization working to advance the knowledge of physics. |
© 2024 American Physical Society
| All rights reserved | Terms of Use
| Contact Us
Headquarters
1 Physics Ellipse, College Park, MD 20740-3844
(301) 209-3200
Editorial Office
100 Motor Pkwy, Suite 110, Hauppauge, NY 11788
(631) 591-4000
Office of Public Affairs
529 14th St NW, Suite 1050, Washington, D.C. 20045-2001
(202) 662-8700