Active 3D weaves for load-bearing and damage-resilient locomotion

Guowei Wayne Tu; Evgueni T. Filipov

cs.ROSep 26, 2026

Active 3D weaves for load-bearing and damage-resilient locomotion

Authors: Guowei Wayne Tu, Evgueni T. Filipov

Organizations: Mechanics and Materials of Bio-Integrated Electronics Laboratory, Department of Aerospace Engineering and Engineering Mechanics, University of Texas, Austin, TX, USA · Reconfigurable, Architected, and Regenerative Structures Laboratory, Department of Civil and Environmental Engineering, University of Michigan, Ann Arbor, MI, USA · Reconfigurable, Architected, and Regenerative Structures Laboratory, Department of Mechanical Engineering, University of Michigan, Ann Arbor, MI, USA

Abstract

The craft of weaving, where different materials are interlaced, has tremendous potential for creating active and functional systems for use in soft robots, prosthetics, wearables, exoskeletons, and more. These textile-like systems are flexible and safe for human-machine interaction; however, this inherent flexibility limits their ability to carry loads, which is essential for many robotic functions. In this work, we introduce a general framework for integrating active materials into three-dimensional (3D) woven shells to create robotic structures that combine high axial stiffness for load bearing, low bending stiffness for efficient actuation, and system-level resilience for damage tolerance. These woven robots can be modularly assembled from 'woven corners', a fundamental unit of 3D woven structures. We use eigenvalue calculations to identify load bearing and actuation mechanisms of the 3D woven structures, and use that information to make five different robots capable of locomotion. We demonstrate that these 3D woven robots can locomote carrying loads 70 times their self-weight, and can maintain repeatable performance even after being subjected to extreme compression. This work is a pathway toward the design, manufacturing, and simulation of future 3D woven robotic systems where load bearing, high stiffness, active functional deformation, locomotion, and system-level resilience are all needed.

Figures & tables

Figure 1: Fabrication and locomotion of 3D woven corners integrated with SMAs. (a) Fabrication sequence from flat ribbons to a 3D woven corner structure combined with SMAs; (b) Actuation of the woven corner structure in (a) where the SMAs are heated through a direct current; (c) and (d) are the control signal and configurations of the woven corner robot in locomotion along a straight line; (e) and (f) are the control signal and configurations of the woven corner robot in steering along a curved line. In (c) and (e), the control current signals for the left and right SMA actuators are given above and below, respectively. In (d) and (f), the images of the moving woven corner are juxtaposed with the simulated and experimental paths (blue and magenta curves); the simulated configurations are given at three discrete times ( $t=$ 0, 395, 800 s), and the images are nine snapshots with the same time duration between each one.

Figure 2: Corner-based modular design of 3D load-bearing woven robots. (a) Woven corner as a fundamental unit for modular assembly of 3D woven robots; (b) and (c) show two types of 3D woven robots assembled from woven corner units that utilize the mechanisms of body-driven locomotion and limb-driven locomotion, respectively. In each column in (b), (c, i ), (c, ii ), and (c, iii ), we show the simulated relaxed (unactuated) and actuated configurations of each woven robotic structure where a corner unit is highlighted; below the simulations, we show the images of each moving woven robot that are juxtaposed with the simulated and experimental locomotion paths (blue and magenta curves), and the images are ten snapshots with the same time duration between each one.

Figure 3: Mechanics modeling of 3D woven robots. (a) The bar & hinge mesh of an active woven corner under external load. A schematic of bar and hinge elements are shown on the right; (b) The four configurations of the loaded woven corner from the bar & hinge simulation: the trivial shape, the rest shape with both actuators off, the actuated shape in locomotion with both actuators on, and the actuated shape in steering with only one actuator on; in the trivial shape, the external load and boundary condition are shown in green; (c) ( i ) and ( ii ) show how we calculate the step size of locomotion and steering of the woven corner robot using the bar & hinge simulation; the left and right columns in ( i ) and ( ii ) show a comparison between experimental and simulated configurations in one actuation cycle. For locomotion along a stright line (c, i ), we use $\Delta$ to represent the translational step size in one actuation cycle; while for steering along a curved line (c, ii ), we use triangles formed by the three vertices to characterize the configurations of the woven corner (the white, blue, and magenta triangles in (c, ii )); we use the magnitude $L_{1}$ and the counterclockwise angle from the axis of symmetry $\theta_{1}$ to represent the translational vector of the center of the triangle in the $x$ – $y$ plane, and we use $\theta_{2}$ to represent the angle of rotation of the axis of symmetry of the triangle in each actuation cycle.

Figure 5: Damage resilience of 3D woven robots. (a) Snapshots of a video where a woven corner robot recovers its initial geometry and continues locomoting after an extreme compression from the external load 1300 times its self-weight. The robot can withstand this extreme compression multiple times and resume locomotion, which is shown in Supplementary Video 2; (b) The experimental setup to compress a woven robotic structure to a height that is 1/4 of its initial height; (c) The consecutively repeated locomotion and steering paths of the woven corner robot without compression; (d) The consecutively repeated locomotion and steering paths of the woven corner robot with compression after each path; the compression is done using the setup in (b); (e) and (f) give the path deviation between the experimental path and the simulated path in the 10th locomotion in (c) and (d); (g) the root-mean-square path deviation of the 10th locomotion of each woven robot without compression and with compression after each compression cycle. The error bars are calculated based on three tests of three samples fabricated the same way, where the upper boundary is the highest value and the lower boundary is the lowest value among all three tests.

Robotics

Robotics studies the design and control of machines that perceive, move, and act in physical environments.

See all 751 papers

A Reduced Order Model for Emergent Mechanics in Woven Systems
Jun 22, 2026Anvay A. Pradhan, Evgueni T. Filipov, Talia Y. MooreStructural Mechanics Materials Science
Department of Mechanical Engineering, University of Michigan, Ann Arbor, Michigan, 48109, USA · Department of Civil and Environmental Engineering, University of Michigan, Ann Arbor, Michigan, 48109, USA · Department of Robotics, University of Michigan, Ann Arbor, Michigan, 48109, USA +1
ZipFold: Modular Actuators for Scaleable Adaptive Robots
Apr 6, 2026Niklas Hagemann, Daniela RusModular Robotics
Computer Science and Artificial Intelligence Laboratory, MIT, Cambridge, MA 02139 USA
Selective Unit-Cell Actuation in Lattice Structures for Distributed Morphology in Soft Robots
Jun 17, 2026Trevor Exley, Altair Coutinho, Lucia BeccaiPneumatic Actuation Soft Robotics
Soft BioRobotics Perception Lab of the Istituto Italiano di Tecnologia (IIT), Genova 16163, Italy

Jun 22, 2026J/K move · Enter open · S save

A Reduced Order Model for Emergent Mechanics in Woven Systems

Anvay A. Pradhan, Evgueni T. Filipov, Talia Y. Moore

Department of Mechanical Engineering, University of Michigan, Ann Arbor, Michigan, 48109, USA · Department of Civil and Environmental Engineering, University of Michigan, Ann Arbor, Michigan, 48109, USA · Department of Robotics, University of Michigan, Ann Arbor, Michigan, 48109, USA+1

Structural Mechanics Materials Science Reduced-Order Modeling Computational Materials Science

Open paper AskPDF arXiv

Woven structures exhibit rich mechanical behaviors including anisotropic stiffness, shear-induced locking, and crimp interchange that emerge purely from the geometric arrangement of individual weavers rather than from constituent material properties. Existing models either homogenize these interactions or resolve them at prohibitive computational cost. We introduce a reduced-order model that bridges this gap by representing individual weaver interactions through a system of nodes and four physically interpretable stiffness elements capturing axial deformation, in-plane uncrimping, inter-weaver shear, and frictional slip. Eigenvalue analysis of the unit cell confirms that the lowest-energy deformation modes correspond directly to known weave-specific phenomena, and that each element is necessary for a complete kinematic and mechanistic description. Element stiffness parameters are calibrated against empirical three-point bending and shear data, achieving agreement within 5% across varied weaver widths and spacings. The validated model is then applied to demonstrate capabilities beyond the reach of continuum approaches including: the emergent Poisson's response arising from crimp interchange, stepwise force reduction during progressive weaver pullout, stress localization under three distinct tearing configurations, and programmable mechanical anisotropy through spatially graded weaver stiffness. The physical transparency and computational efficiency of the framework position it as a practical tool for the analysis and design of woven architected materials with programmable mechanical response.

Abstract

Robotics

Explore similar work

A Reduced Order Model for Emergent Mechanics in Woven Systems

ZipFold: Modular Actuators for Scaleable Adaptive Robots

Selective Unit-Cell Actuation in Lattice Structures for Distributed Morphology in Soft Robots