This demonstration, based on our full paper, presents a physical prototype of a transformable modular system constructed from Snapology origami. By introducing geometric frustration into assemblies of rhombic and regular triangular prismatic modules, we introduce a snap-through transformation from a 2D to 3D structure. The motion is programmed by activating selected out-of-plane fold hinges around a vertex, enabling either smooth or snapping motion. Participants can experience the snapping motion, triggered by in-plane actuation of rhombic modules. The demonstration showcases the design and fabrication of our Snapology and its potential for programmable mechanical metamaterials.
Structural color-color produced by light reflecting nanostructures rather than pigment-is nature's iridescent color palette. In this demo, we present MorphoChrome, a handheld optical device for programmable, real-time, structural color fabrication. Analogous to painting with light, MorphoChrome creates multicolor designs and color-mixing by exposing a commercially available holographic photopolymer film to user-controlled RGB wavelengths, thereby producing reflective nanostructures. We also introduce a resin-based process to integrate the structurally colored film with flexible and rigid objects.
An automated 3DCP toolpath optimization method is developed and demonstrated on a 4m x 4m post-tensioned modular floor structure. The method includes (1) buildability optimization via normal-driven spherical shape analogy, (2) curve number reduction through adaptive Booleaning, (3) "nonstop" globally continuous toolpath with detouring travel lines, and alternative methods of (4) curved slicing and (5) infill generation.
Multi-material 3D printing combines the functional properties of different materials (e.g., mechanical, electrical, color) within a single object that is fabricated without manual assembly. However, this presents sustainability challenges as multi-material objects cannot be easily recycled. Because each material has a different processing temperature, considerable effort must be used to separate them for recycling. We present a computational fabrication technique to generate dissolvable interfaces between different materials in a 3D printed object without affecting the object’s intended use. When the interfaces are dissolved, the object is disassembled to enable recycling of the individual materials. We describe the computational design of these interfaces and demonstrate our technique across multi-material 3D printed objects of varying structural and functional complexity. Our technique unlocks a new approach that enables recycling in multi-material 3D printing without compromising functionality.
We explore the fabrication of curved surfaces by reusing panels extracted from decommissioned wind turbine blades, using cycling pumptracks as a case study. We first present real-world prototypes of pumptrack modules that we manufactured to evaluate the practicality of this reuse scenario and to define the boundary conditions for harvesting blade panels and assembling a track. We then propose an algorithm to optimize the segmentation of a wind turbine blade into quadrilateral panels whose sides fall within a small set of compatible boundaries. These panels form a library of modules that designers can connect side by side to create pumptracks of various lengths and curvatures. Together, these contributions provide a proof-of-concept of how computer-aided design and manufacturing can support circular design through the reuse of curved surfaces.
We demonstrate SplatOverflow, a workflow for asynchronous hardware troubleshooting. SplatOverflow creates a novel boundary object, the SplatOverflow scene, that users reference to communicate about hardware. A scene comprises a 3D Gaussian Splat of the user's hardware registered onto the hardware's CAD model. The splat captures the current state of the hardware, and the registered CAD model acts as a referential anchor for troubleshooting instructions. In this demo, attendees will be able to create and navigate a SplatOverflow scene and explore how it can be used to coordinate complex troubleshooting tasks asynchronously.
Kirigami patterns, unfolding like deployable nets, intrinsically supply the negative-Poisson's-ratio motion demanded by auxetic metamaterials. Current designs are limited by bottom-up, case-specific analyses. We bypass these bottlenecks with a controllable kirigami framework driven by three geometric numbers: a glide ratio t and two global shape parameters (lambda, mu) that fix every auxetic quad panel to mutual similarity. A checkerboard of identically-shaped parallelograms-auxetic quad panels-is inscribed on every second face of a control net, making the diagonal net isogonal and simplifying inverse-design form-finding. Global programming of open/closed states proceeds under local edge-length constraints and isometric optimization, while identical panels guarantee low-cost scalable fabrication. Initial experiments validate the pipeline: similar auxetic quads fully open on the target surface, and their compact planar development reveals a non-trivial kirigami pattern.
Most of the electric and electronic circuits around us are flat, and applying them to diverse 3D surfaces is challenging. This is because (1) printed circuit boards (PCBs) are generally non-stretchable and cannot fit doubly curved surfaces, and (2) folding PCBs into non-planar origami geometries requires enormous time and manual effort. To overcome these limitations, we propose 4D Leaf Circuits, a method to automatically fold complex 3D circuits solely through inkjet printing and conductive leaf transfer. We extended the existing 4D printing method using an inkjet printer and added a primer ink layer. This layerworks as an adhesive for conductive leaf transfer on selected surface areas. Different from previous work on 4D printed circuits, our method does not require an additional screen printing process of conductive ink or any other modification to the printer, which contributes to faster and easier fabrication of 3D objects with circuits. To identify the materials with robust mechanical and electrical properties, we investigated multiple heat-shrinkable films for shape change and conductive leaves for circuits. We also developed an algorithm that generates optimal circuit layouts, avoiding mountain folds and minimizing path length. To demonstrate the feasibility of our method, we built three design examples of an interactive touch-sensing toy, a custom-made mouse, and a portable glow bangle reacting to the user's motion.
While bodies change over time and trends evolve, most garments are produced in fixed sizes and styles that are difficult to adapt or reuse. Alterations can enable limited changes, but they often require sewing and are typically irreversible. We present Refashion, a modular approach to garment design that supports resizing, restyling, repair, and reuse through reversible assembly. Our system introduces a compact set of fabric modules and connectors, an optimization-based method for decomposing garments into reusable modules, and a digital design tool for creating, visualizing, and simulating modular garments. Our results demonstrate that a small, standardized module set can support a wide range of garment forms and enable users to reconfigure garments across sizes and styles while reusing the same physical building blocks.
Transparent materials transmit light without significant scattering or absorption due to total internal reflection. Transparent channels in 3D printed objects follow this theory, functioning similarly to optical fibers by transmitting input light. While prior work enabled basic interactions in custom optical sensors like push and display interactions using photopolymers, complex channeling remains challenging, particularly in desktop Fused Deposition Modeling (FDM) due to the inherent printing discontinuities. We present a framework that enables low-cost desktop fabrication of optical interactive devices using Voronoi-based segmentation of objects for in-place FDM printing of optical channels. It allows uninterrupted light transmission where the embedded channels are aligned in the XY plane for uninterrupted printing. We further explore the use of FDM optical objects in routing and channeling strategies to support applications in displays, sensing, and embodied interactions.
Ceramic-based evaporative cooling is a process that uses hollow ceramic pipes to cool the surrounding environment, serving as passive air conditioning. It is a low-cost, energy and material-efficient technology that offers valuable local-scale opportunities as global temperatures become increasingly unpredictable. However, designing and fabricating ceramic pipes has posed challenges due to a lack of computational fabrication techniques to support this process. We present CeraPiper, an end-to-end fabrication pipeline that integrates a standard clay extruder with a dynamic die capable of changing shape. CeraPiper comes with a CAD tool to design custom structures for evaporative cooling and control the fabrication process accordingly. By enabling real-time adjustment of the cross-sectional profile of the dynamic die during extrusion, CeraPiper unlocks possibilities for both ceramic-based evaporative cooling as well as creative exploration of hollow pipe extrusion. To demonstrate the potential of CeraPiper, we fabricated proof-of-concept cooling units. Our technical evaluation shows that CeraPiper achieved repeatable geometric control over design primitives and produced ceramic assemblies that lowered chamber temperature, elevated relative humidity by similar to 8-10%, and evaporated 1.85 L in 60 h.
When purging 3D printing filament from a nozzle, you may notice a peculiar coiling behavior as the material deposits on a surface. This phenomenon is known as liquid rope coiling (LRC). In this demo, we utilize LRC to enable users to 3D print soft foam augmentations directly on top of existing objects, as well as create pressure sensors to facilitate interactive applications.
We present a programmable self-folding modular robotic chain capable of transforming input 3D geometries into reconfigurable objects. Our system integrates three components: geometric discretization of target volumes, a path-finding and folding algorithm, and the design of modular robotic chain hardware. To validate the approach, we constructed a 12-module robotic chain and demonstrated folding sequences that approximate everyday objects such as a table, stool, and steps. Preliminary results show that the system can transition between 2D and 3D configurations, achieving stable forms under limited load-bearing conditions.
The design of large-scale and topologically complex tensile structures presents a significant challenge, often requiring bespoke software that separates engineering analysis from creative intent. To make this process more approachable, we introduce Ariadne and Theseus(1), a software toolkit that extends our work on differentiable form-finding with FDMremote by enabling more fine-grained control over network sub-components. This allows designers to manage complexity by optimizing for a set of user-defined and composable objectives while ensuring that the structure remains in static equilibrium. The power of this workflow was validated through its use in the design of Janet Echelman's sculpture, Remembering the Future. Our demonstration will showcase this workflow across a range of examples, from simple networks to the complex structure developed for Remembering the Future, contributing to a proven process for realizing highly complex tensile networks.
Textiles are increasingly recognized in architecture and robotics for their adaptability to diverse shapes, low environmental impact, and lightweight character. While flatbed knitting has enabled the production of complex 3D composites and formwork, the low internal strength and geometric limits of knitted fabrics restrict their structural potential. Crochet, by contrast, offers higher tensile strength, greater three-dimensional extensibility, and the ability to generate arbitrary topological surfaces through its variable stitch geometry. However, the craft has yet to be mechanized due to its complexity and dense stitch structure. This project introduces a new framework for robotic crochet, which translates the handicraft into a loom-based fabrication process. A robotic arm equipped with a latch-needle end effector builds stitches from any point in a constrained fabric matrix, supported by a passive yarn tensioning system. This approach enables the reliable automation of crochet for producing intricate three-dimensional morphologies in architectural composites, soft robotics, and biomedical applications.
Makers regularly discuss substitution suggestions for materials, tools, and practices around DIY projects online. However, these suggestions are often lost in the comment sections of projects, which can lead to mistakes being repeated and slower iterative project improvements. To address this, we propose utilizing large language models to identify, collect, and structure substitution suggestions in users’ comments. We prototyped such a workflow using OpenAI’s GPT-4o model. To evaluate its performance, we labeled 4193 comments regarding whether they contain a substitution and what it is about. The workflow successfully identifies substitution suggestions, including the substitute, original, username, and comment ID, from DIY tutorials on Instructables and YouTube and outputs them in a JSON format for further processing. We report the quantitative performance metrics F1, ROUGE-L, and BERTScore, qualitative insights into limitations and benefits, and solutions to reduce the adverse side effects of generative variability. The collected data is provided as supplements.
We present 3Duino, a unified software and hardware platform that enables users to prototype interactive devices without specialized expertise in mechanical design, electronics and programming. With 3Duino, users can assign desired input and output functionalities (e.g., touch input, motion sensing, lighting, or physical actuation) directly to a 3D model. For each specified function, 3Duino automatically generates the necessary internal interactive structures, designed for single-piece 3D printing, minimizing post-processing and ensuring seamless compatibility with the 3Duino hardware. In addition, 3Duino also allows users to define interaction logic using natural language statements through its interface. Based on these statements, the system generates the corresponding control code to run on the hardware. To inform the design of 3Duino, we conducted a formative study to identify key challenges in existing workflows. We then developed and evaluated 3Duino through a user study with 12 participants, which showed that the platform lowers the barrier to prototyping interactive 3D-printed devices, enabling users to create functional, interactive artifacts with ease.
We propose Otto, a comprehensive 2D design tool that provides accessible parametric designing capabilities through multiple interface modalities, including block-based and text-based interactions, specifically designed to introduce parametric design concepts. We created a web-based1, open-source2, end-to-end environment for 2D parametric forms that are fabricated on laser cutters. Users can manipulate designs by directly editing parameters and constraints that drive geometric relationships and dimensional accuracy. These operations translate bidirectionally between visual blocks and text code, allowing users to seamlessly switch editing modes while maintaining design continuity and enabling adaptive parametric creation. We pilot tested our program with a small group of FabLab students, which suggested applications in education and quick prototyping. We aim to create an interactive environment that makes parametric design intuitive through hands-on experimentation.
