Session A30: Morphing Matter: From Soft Robotics to 4D Printing I

Robotic Morphing Matter as Materialized AI

Morphing Matter are physical materials that are transformable, adaptive and autonomous. They are programmable or pre-programmable with inherent sensing, actuating and computational behaviors across scales from the nano to macro. Situating Morphing Matter in the present, It is materialized AI that makes conventional computers disappear and computation weaves itself into the fabric of everyday life - the ultimate dream of ubiquitous computing envisioned by Mark Weiser. These material systems can be leveraged to design soft robots, self-assembling furniture, adaptive fabrics, and self-folding foods. In this talk, Lining presents the recent works in the Morphing Matter Lab, Human-Computer Interaction Institute of School of Computer Science at Carnegie Mellon University and highlights several robotic morphing materials that weave advanced manufacturing, computational tools, and design thinking. Her team believes that the term “robotics” does not only refer to conventional robotic forms and controls but also connects to the artifacts’ abilities to make decisions, adapt, move, and respond to different stimuli. More information from the lab site: https://morphingmatter.cs.cmu.edu/   

Spider-morphs: Designing 3D shapes from multiple tapered elasticæ.

Foldable three-dimensional (3D) structures are important in a wide range of engineering applications. Transforming flat two-dimensional sheets with cuts into 3D structures, or kirigami, has emerged as an exciting manufacturing paradigm. However, achieving a particular 3D shape usually requires multiple materials and/or the application of external stimuli. Here we introduce a design framework for forming approximately axisymmetric 3D structures by harnessing the buckling of multiple tapered elastic sheets (the legs) connected in a central portion (the body). Together this creates a spider-like structure that morphs in 3D: a spider-morph. We design spider-morphs that deform into axisymmetric 3D structures with positive, negative, and variable Gaussian curvature. We conduct both numerical simulations and physical experiments to verify our theoretical approach.   

Reshapable groovy sheets

Most mechanical metamaterials are designed for a single function. But multifunctional metamaterials need to respond distinctly to different kinds of mechanical input. We show that groovy sheets---thin sheets with parallel corrugations---form patterns of snap-through defects and shift into different shapes, depending on how they are actuated. We show that geometric nonlinearities and frustration lie at the heart of the reversible yet multistable behaviour of groovy sheets.   

Under pressure: Mechanics of swelling hydrogels under confinement

Hydrogels are polymer networks that can absorb considerable amounts of water. They are thus promising additives to soil in arid conditions, increasing water retention and decreasing the need for plant irrigation. However, field measurements indicate that confinement in soil alters both the ability of hydrogels to hold water, as well as the properties of the soil itself—and the underlying physical reasons remain unknown. We have developed the ability to directly visualize hydrogel swelling within a three-dimensional porous medium that mimics soil. Using this platform, we quantify how the presence of the solid grains around a hydrogel hinders its ability to swell. By testing different applied loads and sizes of porous media, we show that the deformations of the hydrogel and the medium can be described by a balance between the osmotic swelling pressure of the hydrogel, the local elastic strain energy needed for the hydrogel to swell into the pores, and the frictional interactions holding grains together. Our results thereby provide a general framework by which hydrogel swelling can be understood, potentially improving their ability to help plants survive drought, and informing applications in new settings like oil fields and lab-on-a-chip devices.   

Delicate and Precise Grasping using Kirigami

The ability to precisely and delicately handle deformable, fragile, slippery, or microscale objects remains a significant challenge for robotic grippers. In this work, through a combination of experiments and modeling, we demonstrate how the mechanical actuation of kirigami shells enables both delicate and precise grasping of a wide variety of objects. The design is geometric, so it can be adapted to many material systems, and the geometry was optimized to produce a structure with a high carrying capacity and a low actuation force. We combined individual kirigami grippers in series and in parallel to form gripper arrays capable of grasping slender rods, and simultaneously grasping multiple objects and moving them without changing their relative orientation and position. The kirigami gripper is scalable, lightweight, and can be incorporated with commercially available robotic system to perform delicate and high precision grasping.   

Design and mechanics of complex inflatable networks

The use of compliant materials to accomplish complex movements, made difficult or impossible by rigid materials, has inspired a wide array of soft robots. We have recently introduced “bubble casting”, a novel assembly method that leverages the fluidity of curing silicone elastomers to easily fabricate soft actuators with complex shapes. While liquid, an elastomer is first injected in a tube or tubular mold. An inner void is subsequently sculpted by injecting an elongated bubble into the channels, leaving elastomer only on the channels walls. As the elastomer cures into an elastic solid, gravity passively drains the top part of the channel and lifts the bubble to form the final actuator whose cross-section consists of a thin upper membrane attached to a thick lower beam. Here we explore the mechanical response of our soft actuators: while linear actuators are found to curl when inflated owing the asymmetry of their cross section, the deformation of closed shapes, such as loops, or that of connected networks is more intricate. We will discuss the experimental results we obtained with these programmable robots, and the models we have derived to rationalize our observations.   

Buckling and Metastability in 2D Impurity Arrays

We study a periodic array of impurities that produce local dilations, embedded in a two-dimensional crystalline solid that can buckle out of the plane. These arrays provide a simple elastic model of shape memory. As the size of each impurity increases (or the relative cost of bending to stretching decreases), it becomes energetically favorable for the impurities to buckle either up or down, allowing for a vast number of metastable states. Using discrete simulations and continuum theory, we consider the buckling of isolated impurities as well as impurity arrays, guided by an analogy to the Ising antiferromagnet. We characterize the buckling transition and conjecture ground states for systems with triangular and square lattice microstructures.   

4D printing of mechanically programmable shape-shifting liquid crystal elastomers

4D printing is a promising method to develop actuators for applications in soft-robotics and biomedical devices where complex structures are required that might be difficult to create using traditional fabrication methods. However, most 4D printing relies on shear alignment upon printing to align liquid crystal or composite fibers that undergo anisotropic expansions/contraction when actuated. This limits the types of shape changes available to researchers as determining the print path required to induce a desired shape change is not trivial. Here, we use a new reactive printing method that enables the printing of a dual network liquid crystal elastomer (LCE) which can be mechanically programmed into the desired shape change. First, a thiol-acrylate Michael addition is completed upon printing an LCE oligomer solution into a catalyst bath. Next, the printed structure is dried, deformed to a desired shape change, and UV cured to crosslink excess acrylates in the network. The resulting LCE transforms between the printed and mechanically deformed shape when heated and cooled, respectively, and is capable reversible strains up to 100%. We demonstrate the versatility of this method by printing a variety of LCE actuators which could not be printed using conventional 4D printing methods.   

Versatile and controllable shape morphing using twisted-and-coiled actuators

Various shape morphing strategies have been investigated in recent years. Even though existing work has demonstrated complicated shape morphing (e.g., a human face), they can only deform along with a predefined pattern. We propose a different strategy: strategically embedding soft actuators into a soft body to enable versatile shape morphing by actively controlling the deformation of each soft actuator. We utilize a soft actuator, twisted-and-coiled actuators (TCAs), which can be easily fabricated from low-cost sewing threads to generate large force and displacement. We also leverage a jamming-based method to change the stiffness of soft bodies. With these two elements, we realize a shape morphing module that can quickly transform to and hold versatile shapes to adapt to various environments. Different modules can also be combined to generate more complicated shapes. We demonstrate a fast grasping and holding process, which consumes no energy, with a gripper with two shape morphing fingers. We also demonstrate a morphing surface that can hold different 3D configurations. We envision that the concept can be applied to soft robots, medical robots, morphing structures, etc.   

Controlled Shape-morphing of Elastic Sheets by Chemically-driven Fluid Flow

Shape-morphing of two-dimensional (2D) materials into complex three-dimensional (3D) structures provides a wide range of applications from wearable electronics to soft robotics. Convective flows generated by the appropriate chemical reactions provide a mechanism for shape transformation of elastic 2D materials submerged in a flow environment. Here, using a computational model that incorporates interrelated chemical, hydrodynamic and mechanical interactions, we demonstrate how a (2D) enzyme-coated elastic sheet can spontaneously morph into 3D structures in response to specific chemical stimuli introduced in the fluid-filled microchamber. We outline design principles for creating a multi-responsive elastic sheet that self-morphs into different 3D structures. We further develop a theoretical model based on lubrication theory for thin liquid films coupled to the deformations of an elastic sheet that rationalizes our simulation observations and provides insights into the initial dynamics of the transformation of elastic sheets within the microfluidic system.   

Morphing Surfaces formed by Liquid Crystal Elastomer Coatings: Design and Modeling

The suction cups on the arms of an octopus are morphing structures that can actuate to adhere to a non-porous surface, and then release on command. To mimic this functionality in an engineered material, we design and model dynamically morphing surface coatings composed of stimuli-responsive liquid crystal elastomers (LCE). In these programmable materials, patterned molecular orientation gives rise to complex shape transformations, driven by change of temperature. Using a combination of Finite Element Modeling (FEM) and analytical calculations, we design and characterize director patterns that create a variety of surface topographies: a lattice of spikes and/or indentations, an array of parallel microchannels or ridges, and patterned zones of positive and negative Gaussian curvature. To drive a thin film coating to morph into a lattice of suction cup-like indentations, we design an LCE with an array of topological defects, where the orientation of the director near each defect core follows a sigmoid function. By tuning adjustable parameters that define the structure of the defect core, we control the resulting surface topography.
  

Active rapid morphing mechanism of the Venus flytrap

The Venus flytrap (Dionaea muscipula) exhibits rapid snapping in about 100 ms, which has long fascinated scientists as one of the fastest botanical movements. Its motion is mechanically initiated by stimulating the trigger hairs, and accelerated by using the snap buckling instability of the poroelastic curved shell geometry of the leaves. At a macroscopic level, its kinematics and dynamics are well understood. However, the mechanism to actively change its natural curvature remains unknown. Here we first characterise the ‘active’ dynamics of the Venus flytrap by removing the buckling instability. We probe the change of its mechanical properties before and after triggering the closure and investigate the key components driving this rapid motion at a microscopic level. Then we elucidate the physical mechanisms underlying the rapid morphing of the Venus flytrap. A better understanding of active movements in plants could allow us to design new rapid and programmable morphing structures.