CS Paper Radar

Physically Interpretable Basin Selection in Discrete-Variable Origami Design

Lucy Liu, Leon M. Kamp, Katia Bertoldi, L. Mahadevan

cs.LG cond-mat.mtrl-scicond-mat.soft ·submitted 2026-11-12 17:59 UTC · v1

Abstract

Automatic differentiation has enabled rapid progress on continuous-variable mechanical design problems, but gradient-based optimization still struggles with discrete variables. The standard approach is to optimize using a continuous relaxation of the choices, then commit to the nearest discrete design at the end. Why this procedure selects unfavorable basins is poorly understood. Here, we introduce a practical, interpretable design problem inspired by origami, in which automatic differentiation is used to design shapes with a target functional behavior. Each origami sheet has a fixed geometry, and a bilevel optimization framework tunes which one of four discrete hinge types to place at each fold. Baseline gradient-based design settles into basins with a large discretization gap: relaxed optimization solves 97% of runs, but the final, discrete solutions succeed only half the time. We find that the gap can be relieved with a surprisingly simple modification. An initialization prior uniformly biased toward the stiff, flat hinge option boosts the final solve fraction from 49% to 69%. Applying the prior adds no additional compute and leaves the optimization process unchanged. Compared to the baseline, sheets produced with the prior have greater geometric compatibility and mechanical stiffness. Directly penalizing frustration and floppiness in the outer loss function produces an even greater improvement than the prior does, raising the solve fraction to 78% and indicating that these properties help distinguish the unfavorable basins. Our results demonstrate that loss landscapes for mechanical design can have physically meaningful basins that are not robust to discretization and cause solves to fail. Targeting these basins' physical signatures, even with just an initialization bias, helps avoid them.

Closest papers

Nearest neighbours by meaning across the whole index.

Shape optimisation of nonlinear Naghdi shells on discrete geometries

cs.CE math.NA ·2026-08-11 ·Ado Farsi, Alberto Paganini
benchmark similarity 0.58
Abstract · 19 pages, 10 figures, 6 tables. Revised manuscript submitted to Computers & Structures

We present an automated framework for the shape optimisation of a geometrically nonlinear, five-parameter Naghdi model on faceted triangulations. By recovering directors numerically, we remove the requirement for an analytic or spline parameterisation of the mid-surface, whereas partial selective reduced integration alleviates locking. The residual and tangent of symbolic energies are generated in an automated fashion building on Firedrake's and UFL's symbolic differentiation capabilities. Additionally, UFL's and pyadjoint's algorithmic differentiation enables automated first and second shape derivative evaluations that are verifiably correct via Taylor tests. The forward model agrees with the Sze/Abaqus semi-cylinder reference. A three-model mesh-refinement study indicates that geometry recovery and surface faceting have small effects on the predicted deflection. Numerical experiments show that 100 trust-region iterations suffice to reduce the squared-displacement functional of a semi-cylinder by 98.4% and the elastic strain energy of a bracket by 94.6%. The latter result is validated against a COMSOL benchmark. Non-uniqueness of optimal shapes is demonstrated by optimising initial designs subject to small perturbations. Comparisons with linearised designs distinguish optimisation performance from the accuracy of the predicted nonlinear response.

0.12

Inverse origami inspired by insect morphogenesis: geometric design of wrinkle structures for curved surface formation

cs.GR cond-mat.softphysics.app-phphysics.bio-ph ·2026-10-26 ·Kentaro Morikawa, Yuuki Shimizu, Nen Saito, Yasuhiro Inoue
unusual similarity 0.55
Abstract · 10 pages, 7 figures; SI Appendix: 4 pages

Insect exoskeletons abruptly acquire complex three-dimensional forms during metamorphosis from larva to pupa. This transformation occurs through the unfolding of preformed wrinkle structures inside the larval body under internal pressure. Inspired by this process, we investigate an inverse-origami strategy for fabricating curved surfaces. Whereas conventional origami produces three-dimensional objects by folding a flat sheet, this strategy produces curved surfaces by unfolding preformed wrinkle structures. The central design problem is to derive a wrinkle structure from a prescribed target surface. Our key idea is to regard a wrinkle structure as a unidirectional stretching field, which specifies at each point the stretch produced when the structure unfolds. We first derive such a field from the target surface geometry and then translate it into spatially varying wrinkle directions and spacing, so that the wrinkle structure unfolds into the target. The resulting numerical designs approximate the intrinsic metric of the target surface. We demonstrate the method on multiple target surfaces through numerical construction, mechanical unfolding simulations, and 3D-printed prototypes. This work provides a biologically inspired geometric design principle for encoding prescribed surface shapes in wrinkle structures, as well as a framework for designing and fabricating deployable curved surfaces.

0.35

Robust Optimization of Spring Design under Variable Manufacturing Tolerances

cs.NE ·2026-10-03 ·Grzegorz Sroka, Slawomir T. Wierzchon
similarity 0.54
Abstract

A reliable assessment of an optimizer robustness requires tracking controlled changes in parameters to identify sources of algorithmic weakness. Such an approach was applied to the problem of designing tension/compression springs with three continuous variables describing the spring geometry. Three additional independent binary parameters reflect material variability, manufacturing-related geometric deviations, and an additional safety margin. Without expanding the decision space, these parameters can be combined in eight ways to generate variants that allow for direct comparison. In this study we distinguish between robustness effects and the dimension of the problem, which allows tracking changes in feasibility while keeping the nominal mass objective constant. Seven optimization methods and eight hybridized their variants are used for the analysis. The hybridization mechanism applies the operator selected by the prediction procedure presented in Section 3. All methods are evaluated under the same experimental conditions. Each method is run thirty times for each variant, with a maximum of 30000 function evaluations permitted per run. The final quality assessment takes into account the geometry of the feasibility region, search trajectories, performance variations across landscapes and the computational cost of fixed objectives. Together, these factors allow us to determine the quality of the solution, reproducibility, feasibility control, and the effectiveness of optimization. An interesting observation was that hybridization can mitigate cost or feasibility losses associated with a specific configuration. While it compensates for a specific weakness in the underlying mechanism, it does not provide universal resilience.

0.45

A Three-Parameter Binary Subdivision Scheme for Shape-Controlled Curve Design

cs.GR ·2026-08-23 ·Rabia Hameed, Jihad Younis, Maryam Salem Alatawi, Tahreem Farman, Hafiza Sana Mukhtiar
similarity 0.53
Abstract · This is only a part of the article, not the complete article. 14 pages 2 Figures

Shape-controlled curve design plays a fundamental role in computer-aided geometric design, computer graphics, and engineering applications. In this paper, we present a novel three-parameter 9-point binary approximating subdivision scheme constructed by a weighted combination of the refinement rules of the 7-point Lagrange and 7-point B-spline subdivision schemes. The proposed construction employs displacement vectors between the corresponding refinement points of the parent schemes and constructs resultant vectors by combining neighbouring displacement vectors through three independent design-control parameters. These resultant vectors are subsequently used to derive the refinement rules, thereby yielding a unified family of binary subdivision schemes with adjustable geometric characteristics while preserving the approximating nature of the refinement process. Two representative sub-schemes are derived as special cases by imposing suitable constraints on the design-control parameters. Theoretical investigations establish the support, continuity, endpoint rules for open polygons, and Gibbs oscillation behavior of the proposed family, while graphical examples demonstrate the influence of the design-control parameters on the generated limit curves. Owing to its flexible geometric construction, intuitive parameterization, and favorable mathematical properties, the proposed subdivision framework provides a flexible and effective framework for design-controlled curve design and a valuable tool for computer-aided geometric design and related engineering applications.

0.32

From Biological Curves to Soft Modular Robots: Learning-Based Inverse Design with Uncertainty-Guided Refinement

cs.RO cs.LG ·2026-10-27 ·Dixiang Wang, Minghao Zha, Arifur Rahman, Yinhong Qin, Yitao Jiang +4 more
similarity 0.53
Abstract · 8 pages, 8 figures, 2 tables

Animals exhibit diverse shape-changing behaviors, from the undulatory swimming of eels to the arching movements of inchworms. Reproducing these biological shapes with modular soft robots offers a promising approach to developing versatile bioinspired robotic systems. However, translating a desired shape into a feasible robotic configuration remains challenging. Existing cable-driven modular soft robots rely on computationally expensive physics-based modeling and combinatorial search for inverse design, making it difficult to efficiently reproduce diverse target shapes. To address this challenge, we present a learning-based real-to-sim-to-real framework that translates biological curves into robotic configurations through surrogate-assisted inverse design and uncertainty-guided refinement. The proposed method combines learned inverse prediction with targeted candidate exploration, using a fast forward surrogate to evaluate and select suitable designs. The learned forward surrogate reduces deformation prediction time from $1.53 \times 10^5$ ms to 0.0008 ms, achieving approximately a $1.9 \times 10^8$-fold speedup. At the same time, the proposed inverse design method reduces design time from $1.99 \times 10^6$ ms to 4.18 ms. On 5,000 test samples generated from previously unseen threading patterns, the method achieves a 93% shape reconstruction success rate under surrogate-based evaluation. Physical demonstrations with four bioinspired robots further illustrate the translation of biological curves into realizable robotic shapes.

0.34

Learning Feasibility-Aware Latent Spaces for Preference-Based Exploration of Procedural Automotive Wheel Designs

cs.HC ·2026-09-01 ·Takashi Owaki, Yuki Koyama, Tomoyasu Nakano, Takahiro Yamaguchi, Masataka Goto +1 more
similarity 0.53
Abstract

Intelligent design interfaces that rely on preference-based optimization are most useful when their suggestions are both meaningful to users and feasible within the target domain. Procedural models offer compact and editable design spaces, but their native parameters can be entangled and can generate many invalid outputs, causing human-in-the-loop optimizers to waste comparisons. We propose an interaction-oriented representation-learning pipeline for procedural models and study it in automotive wheel design. The method first screens procedurally generated samples using geometric rules and finite-element analysis, then learns a reduced latent space from the screened subset. We further introduce supervised functional alignment, which reserves selected latent dimensions for stiffness, strength-related stress response, or weight so that search can be biased toward functionally meaningful regions. Simulation experiments show that screened reduction improves target-shape retrieval and the feasibility rate of suggestions, whereas unscreened reduction degrades both. Additional simulations show that constraining search along learned functional dimensions accelerates exploration toward target functional properties. A controlled study with 40 participants further shows that a 5D feasibility-aware space yields higher shape similarity and more feasible suggestions than the original 9D procedural parameterization. These results suggest that, for intelligent user interfaces in engineering design, the representation exposed to the user is a central part of the interaction design, not merely a preprocessing step for the optimizer.

0.27

Rapid Kirigami Simulation using the Bar & Hinge Approach

cs.CE cond-mat.other ·2026-08-13 ·Raj Pradip Khawale, Elaheh Mehdizadeh, John Brigham, Evgueni T. Filipov
published similarity 0.52
Abstract · 23 pages, 11 figures, journal paper

Kirigami, the art of cutting sheets, offers unique characteristics such as shape morphing, stretchability, and adaptability, with applications in deployable and reconfigurable structures. While Finite Element (FE) approaches are widely used to analyze kirigami structures, they are computationally intensive and prone to convergence issues. On the other hand, existing simplified and theoretical models for evaluating kirigami are typically restricted to specific designs and analytical scenarios. This paper proposes a generalized, computationally efficient reduced-order model based on the bar and hinge approach for analyzing any kirigami system. The model represents the entire kirigami structure using truss bars, bending hinges, and torsional springs, capturing both the deformation and internal forces of the system. We derive stiffness expressions for kirigami structures that experience stretching and out-of-plane bending deformations, as well as for structures that experience only in-plane stretching and rotations. The accuracy and robustness of the model are validated through comparisons with FE analysis and experimental data on both single-cut kirigami designs and more complex designs with multiple cuts. Our results demonstrate less than 5% difference in deformation predictions and achieve at least a tenfold computational speedup compared to FE simulations. Additionally, we demonstrate the model's capabilities through simulating complex scenarios, including kirigami-skinned crawlers, high-throughput property-space exploration, and in-plane kirigami-inspired metamaterials. These examples demonstrate computational capabilities that enable large-deformation structural simulations and high-throughput design exploration beyond the practical limits of conventional FE methods.

0.27

Robot Tool Design from Scratch via Behavior-Aware Hierarchical Optimization

cs.RO ·2026-09-28 ·Yinghan Chen, Xiyao Tian, Yizan Dai, Yuyang Li, Yixin Zhu
similarity 0.52
Abstract

The ability to design a tool for a task marks a level of intelligence beyond merely understanding, selecting, or using one. Existing methods for robotic tool design typically optimize a tool's continuous shape and action within a structure that is prescribed or generated beforehand, so the structure itself stays outside the physical optimization loop. We study task-driven tool design from scratch, where tool structure, shape, and action are all derived from the desired physical outcome. Here we show that the three elements can be designed jointly by HOT, a hierarchical optimization whose upper level searches over discrete tool structures with BASS, while lower-level physical optimization evaluates their task behavior and returns milestone progress as behavioral evidence for the search, ultimately providing jointly optimized shape and action. On four tool-use tasks with distinct physical functions, HOT discovers functional structures after evaluating only a small fraction of search spaces containing up to 56 million structures, and the subsequent refinement of their geometry lowers the task loss on all tasks while preserving success, through deformations that are functionally interpretable. Once 3D printed, the tools accomplish all tasks on a real robot with the actions found in simulation. Designing tools from required physical effects, rather than a catalog of known tools, is a step toward the open-ended tool making seen in humans and animals.

0.23