SCF '23: Proceedings of the 8th ACM Symposium on Computational Fabrication

Computational fabrication relies on time-synchronized operation of various machine components. Designing machines for novel workflows is of interest to the computational fabrication community, but designing control systems for these machines, especially with diverse actuators and sensors, remains challenging. We present MAXL, a modular, extensible machine control architecture that enables synchronous control of heterogeneous components. We contribute (1) a design pattern for a distributed trajectory object with one author and multiple readers, (2) high- and low-level APIs for interfacing this trajectory object to modular hardware and to digital fab applications (3) a simple time-synchronization algorithm and queuing scheme for distributing the trajectory object, and (4) an extensible hardware implementation of MAXL. We demonstrate MAXL’s utility in developing new computational fabrication applications by integrating it into two motion control applications; one for time-synchronized data output (light-painting), and the other for time-synchronized data retrieval (from an accelerometer). Finally, we discuss how MAXL can be extended for use in future machine applications.

Recently, soft and deformable materials have become popular as sensors for their applicability in daily objects. Although studies have been conducted on existing conductive soft materials, problems such as a lack of design freedom regarding softness, shape, and deformation, as well as wiring complexity remain. Here, we propose a novel soft sensor called LattiSense, fabricated using an FDM 3D printer. By arranging conductive and non-conductive flexible filaments in a lattice structure, we created a soft sensor designed with a high degree of freedom in terms of hardness, shape, deformation, and wiring paths. By modifying the lattice parameters, multiple modes of deformation can be designed. The softness can also be locally customized by adjusting lattice parameters. In this paper, we present the design and implementation of LattiSense and investigate its characteristics with respect to several parameters. We also demonstrate design software and several application scenarios.

We show that a linear model is sufficient to accurately estimate the quantity of yarn that goes into a knitted item produced on an automated knitting machine. Knitted fabrics are complex structures, yet their diverse properties arise from the arrangement of a small number of discrete, additive operations. One can estimate the masses of each of these basic yarn additions using linear regression and, in turn, use these masses to estimate the overall quantity (and local distribution) of yarn within any knitted fabric. Our proposed linear model achieves low error on a range of fabrics and generalizes to different yarns and stitch sizes. This paves the way for applications where having a known yarn distribution is important for accuracy (e.g., simulation) or cost estimation (e.g., design).

In fused filament fabrication (FFF), the orientation of a part within the printer volume can dramatically affect print quality and probability of success. An object’s orientation determines how much support structure will be required and the strength of adhesion between the deposited material and the build surface. Selecting a part’s orientation is a non-trivial problem that users of FFF slicing software face routinely. Numerous part orientations need to be considered to find the best according to the results of the slicing process. This paper presents a method to automatically determine an optimal printing orientation for FFF that maximizes build-surface adhesion while minimizing the need for support structure. The algorithm considers the slicing angle and a configurable angle for overhang that requires supporting structure. By employing GPU acceleration and convex hull analysis to limit candidate orientations, the algorithm can run in real time as a preprocessing aid to users slicing parts.

It is common to manufacture an object by decomposing it into parts that can be assembled. This decomposition is often required by size limits of the machine, the complex structure of the shape, etc.To make it possible to easily assemble the final object, it is often desirable to design geometry that enables robust connections between the subcomponents. In this project, we study the task of dovetail-joint shape optimization for stiffness using gradient-based optimization. This optimization requires a differentiable simulator that is capable of modeling the contact between the two parts of a joint, making it possible to reason about the gradient of the stiffness with respect to shape parameters. Our simulation approach uses a penalty method that alternates between optimizing each side of the joint, using the adjoint method to compute gradients. We test our method by optimizing the joint shapes in three different joint shape spaces, and evaluate optimized joint shapes in both simulation and real-world tests. The experiments show that optimized joint shapes achieve higher stiffness, both synthetically and in real-world tests.

We introduce StructCode, a technique to store machine-readable data in laser-cut objects using their fabrication artifacts. StructCode modifies the lengths of laser-cut finger joints and/or living hinges to represent bits of information without introducing additional parts or materials. We demonstrate StructCode through use cases for augmenting laser-cut objects with data such as labels, instructions, and narration. We present and evaluate a tag decoding pipeline that is robust to various backgrounds, viewing angles, and wood types. In our mechanical evaluation, we show that StructCodes preserve the structural integrity of laser-cut objects.

We present WireShape, a CNC-supported semi-manual rapid-proto-typing process to create robust inflatable objects. The resulting air chamber systems provide a basis for the construction of shape-changing and malleable user interfaces by integrating sensors, or to create soft robotic actuators. Our process leverages CNC-fabricated bent-wire shapes as custom heat sealing tools for reliable fusing of thermoplastic films or coated textiles. The tools can be reused multiple times and the heat sealing time for fabrication of an individual inflatable is independent of its size (e.g. the length of the two-dimensional welding path), which enables time efficient prototyping of high-quality inflatable structures in small series and at relatively low-cost. We demonstrate a custom CNC-machine for wire bending, a vector graphics interpreter to generate the machine specific CNC-data (G-Code) from 2D vector drawings and outline our fabrication toolchain. The capabilities and limitations of our approach are discussed and preliminary observations and fabrication results from initial workshop sessions are provided.

In 3D printing, stiff fibers (e.g., carbon fiber) can reinforce thermoplastic polymers with limited stiffness. However, existing commercial digital manufacturing software only provides a few simple fiber layout algorithms, which solely use the geometry of the shape. In this work, we build an automated fiber path planning algorithm that maximizes the stiffness of a 3D print given specified external loads. We formalize this as an optimization problem: an objective function is designed to measure the stiffness of the object while regularizing certain properties of fiber paths (e.g., smoothness). To initialize each fiber path, we use finite element analysis to calculate the stress field on the object and greedily “walk” in the direction of the stress field. We then apply a gradient-based optimization algorithm that uses the adjoint method to calculate the gradient of stiffness with respect to fiber layout. We compare our approach, in both simulation and real-world experiments, to three baselines: (1) concentric fiber rings generated by Eiger, a leading digital manufacturing software package developed by Markforged, (2) greedy extraction on the simulated stress field (i.e., our method without optimization), and (3) the greedy algorithm on a fiber orientation field calculated by smoothing the simulated stress fields. The results show that objects with fiber paths generated by our algorithm achieve greater stiffness while using less fiber than the baselines—our algorithm improves the Pareto frontier of object stiffness as a function of fiber usage. Ablation studies show that the smoothing regularizer is needed for feasible fiber paths and stability of optimization, and multi-resolution optimization helps reduce the running time compared to single-resolution optimization.

We present a playful 3D modeling tool that integrates a visual grammar with shaping curves to model forms as parametric functions which can be created and manipulated by novice users. We created a web-based1, open-source2, end-to-end design environment for 3D printable forms which were fabricated on clay 3D printers. Users can manipulate forms by directly editing curves which drive transformations to the form’s profile as it grows in the z-direction. These transformations can be stacked and composed in a visual postfix language which allows for the creation of parametric designs driven by directly manipulable curves. We pilot tested our program with a small group of ceramicists, artists, and digital fabrication practitioners, which suggested applications in education and art making. PotScript demonstrates an approach to integrating direct manipulation into parametric design which allowed users to create complex forms in small sets of user actions. We aim to make modeling complex 3D forms as approachable as stacking a handful of blocks.

We present History in Motion (HiM), an interactive visualization tool that enables CAD designers to interactively explore the design history of 3D CAD models. In contrast to manually exploring the modeling history of a CAD project, HiM finds relevant modeling features for geometry elements selected by the designer. We contribute a novel 3D interactive animation that visualizes how the modeling features interact, and are used on top of the CAD model, to realize the geometry. A control panel allows for a deeper exploration of the modeling features, with shortcuts for making modifications.

While resin 3D printing affords designers unprecedented geometric complexity, currently the technology requires cumbersome support structures that are not user-friendly in numerous ways: materially wasteful, human labor-intensive, time-consuming to remove, damaging to surface finish, and often unreliable in ensuring printability in the first place. Inspired by recent physical demonstrations that injecting liquid through an embedded channel can neutralize suction during printing, we discover this phenomenon can also significantly reduce the quantity of supports required by a user’s model, or entirely obviate them for certain product geometries. We show our fluid dynamics simulation-guided design tool, applied to an off-the-shelf 3D printer with an inexpensive syringe pump add-on, enables designers to more quickly fabricate otherwise unprintable overhang geometries without support.

In construction robotics, a conventional design-to-fabrication workflow starts with designing a structure, followed by task and robotic motion planning, and ultimately, fabrication. However, this approach can prove unsuccessful, as we may only discover the infeasibility of a design at the final stages of the process. This can result in rework and a considerable waste of time and resources. To overcome this challenge, we propose a design method based on reinforcement learning (RL) where the agent makes decisions at every step of the sequential assembly of the structure while considering assembly’s stability. In this way, we take the construction constraints into consideration at the design stage. The research particularly focuses on the design of spanning structures that multiple robot arms can construct without the need for scaffolding.

A series of experiments were conducted using both a centralized and a decentralized learning setup. Our results show that while the decentralized setup was successful in constructing smaller structures, only the centralized setup allowed active collaboration between robot arms, resulting in structures with larger spans. To validate our approach, we fabricated two of the designed structures with two collaborating robot arms, which confirmed the feasibility of these designs. In summary, the proposed method opens exciting possibilities for generating innovative designs that push the boundaries of architectural creativity while simultaneously fulfilling fabrication-related constraints.

We demonstrate the initial results of our exploratory work, investigating the potential for creative machines to collaborate with people through movement as a form of implicit interaction. The paper describes a Wizard-of-Oz demo, where a hidden wizard controls an AxiDraw drawing robot while a participant collaborates with it to draw a custom postcard. This demonstration aims to gather perspectives from the computational fabrication community regarding how practitioners of fabrication with machines experience interacting with a mixed-initiative collaborative machine.

80% of women are wearing an incorrect bra size due to the current limitations in sizing, shape, and fitting methods of bras. Custom-made bras could eliminate these fit issues, alleviating pain and discomfort while boosting women’s confidence. We conducted a study using 3D scanning technology available on iPhones, where we had eight women take a scan of their breasts. From these scans, we extracted measurements and used them to generate custom digital bra patterns. Subsequently, we laser-cut the pattern pieces and sewed each woman a custom bra. Overall, women reported a superior fit with the custom bra but identified areas where further customization was needed.

History in Motion (HiM) is an interactive visualization tool that enables CAD designers to interactively explore the design history of 3D CAD models. In contrast to manually exploring the modeling history of a CAD project, designers can select geometry elements to find relevant modeling features in HiM. These modeling features are then explained to designers using a novel 3D interactive animation that visualizes how the modeling features interact, and are used on top of the CAD model, to realize the selected geometry. A control panel in HiM allows for a deeper exploration of the modeling features, with shortcuts for making modifications. During this demonstration, attendees can experiment with HiM on a variety of CAD designs and explore their design histories.

Folding mechanisms such as origami, linkages, or auxetic materials have found applications across a diverse range of fields, including architecture, product design, and robotics. On the other hand, waffle structure, namely planar section structure, is known as a type of static structure with stable load-bearing capabilities. In this paper, we transform these waffle structures into a foldable mechanism. We introduce "Fold It", a computational design tool that converts arbitrary closed brep or mesh into adaptable waffle structures with tunable folding parameters. Our tool democratizes the personal fabrication of these structures, facilitating the design of diverse everyday objects, ranging from toys to furniture. We hope Fold It could be used for creating more multifunctional objects within and beyond the DIY and personal fabrication arenas.

3D printed objects produced by fused deposition modeling have functionally and aesthetically unpleasing surfaces due to factors inherent to the process. Although it is possible to reduce the resolution of these artifacts during printing, it is not possible to eliminate them entirely, and thus postprocessing will be required to achieve a smooth surface finish. Many methods of postprocessing require a large amount of time and labor or chemical solvents. Using an unmodified laser cutter as a source of heat, we are able to achieve a great improvement in surface finish with minimal warping and fast cycle time. We propose laser polishing as a method that is simple, safe, low-cost, and uses readily available equipment.

Different unmanned aerial system (UAS) applications, such as military reconnaissance, environmental monitoring, or commercial delivery, require different specifications such as payload weight and travel range, which typically require completely different designs. This means that it is difficult, if not impossible, to easily reconfigure an existing UAS for a new application without the redesign and reconstruction of the system. To facilitate easy reconfiguration of UAS, we present Voxelcopter, a modular construction kit for the rapid-prototyping and rapid assembly of multicopter UAS. We develop basic viable components for flight, including reconfigurable building blocks, termed functional voxels, that assemble into high-stiffness, lightweight structures with integrated power and data, a custom flight controller software library, propellers, and power storage. In this demo, a quadcopter is assembled, and we flight test the resulting vehicle, then disassemble it into components ready for reuse, demonstrating the easy reconfigurability of the system.

Models printed using Fused Deposition Modeling (FDM) are characterized by their layered texture, created by the production process. In recent years more focus has been paid towards creating a customized and controllable texture. Controlling the texture can be done using CAD, slicing software or toolpath manipulation. In this early work, we present a method for generating a toolpath that fabricates a drop-like texture for FDM using PLA and TPU materials. The developed texture would be difficult to achieve using other methods. We demonstrate the potential of the drops to provide structural strength, haptic surfaces, and lenticular effects.

Titanium anodization produces a surface coating whose color is dependent on the applied voltage. This process is used in both commercial and hobbyist contexts to create a solid or gradient coloring by dipping titanium into electrified electrolyte baths. Line-drawing by hand is also possible, but the high voltage required for the full range of colors (0-110V) presents significant safety issues. In this paper, we present a computational numerical control (CNC) machine and accompanying software to draw with titanium anodization, providing a safe, efficient, and precise alternative to hand-drawing. We also introduce a technical classification of anodized titanium drawing techniques and a new "inkblot" technique for producing radial gradients.

We demonstrate a technique for constructing smooth, manifold surfaces from flat material. We use this technique to build a Gyroid, a periodic minimal surface, from 121 unique laser cut aluminum panels. The process involves breaking the surface up into panels, perforating those panels with a sparse patterns, and then conformally flattening the panels. The 3D surface is then reconstructed by joining the panels with fasteners at precise connection points. Our technique requires no forming or fixtures, so it works with materials that cannot stretch like wood or paper and is limited only by the curvature of the surface and the ability of the material to bend without creasing.

In this demo, we present an open source methodology for turning triangle meshes into biologically inspired 3D printable objects using an algorithm based on the behavior of the Physarum polycephalum slime mold. Users can import watertight 3D models and have full control of the simulation, allowing for granular control over the density and weight of the resulting network. This is the first open source tool to use this technique that is intended for public consumption and use.

Recently, soft and deformable materials have become popular as sensors for their applicability in daily objects. Although studies have been conducted on existing conductive soft materials, problems such as a lack of design freedom regarding softness, shape, and deformation, as well as wiring complexity remain. Here, we propose a novel soft sensor called LattiSense, fabricated using an FDM 3D printer. By arranging conductive and non-conductive flexible filaments in a lattice structure, we created a soft sensor designed with a high degree of freedom in terms of hardness, shape, deformation, and wiring paths. By modifying the lattice parameters, multiple modes of deformation can be designed. The softness can also be locally customized by adjusting lattice parameters. In this paper, we present the design and implementation of LattiSense and investigate its characteristics with respect to several parameters. We also demonstrate design software and several application scenarios.

Traditional slicing for 3D printing involves a planar cross-sectioning process to create layers. These layers are stacked one on top of the next representing a butt-joint, which is typically the weakest direction for loading a 3D printed part. Because the layers are printed sequentially, no material or reinforcement connects from one layer to the next. This clean layer line between layers is easy to break or separate. To fix this, the authors have implemented a new approach to shift the height of every other closed-loop contour toolpath such that a clean layer line no longer exists. This bead shifting approach involves moving every other contour by half a layer height, after the cross-sectioning step, so that a straight line cannot be drawn between layers. To maintain the original object geometry, a half height layer is printed for the first and last layer to create a flat top and bottom surface. This functionality has been implemented and tested as part of ORNL Slicer 2, an open-source toolpathing software package.

SquiggleDraw is a generative art project that explores documentation techniques for digital fabrication workflows. Participants are invited to make a scribble drawing based on the scribbles’ qualities: line length, turn radius, point compression; then, they watch as a 2D pen plotter recreate their scribble. The generative art is deterministic – the same parameters yield the same drawing – and the parameters to recreate their scribble are printed on a thermal receipt printer. The choice to document user parameters on a receipt paper is intentional; the physical receipt can be kept alongside the fabricated object to recall settings and track outcomes when testing different parameters.