Low-distortion planar embedding of rod-based structures

Rod-based structures are commonly used in practical applications in science and engineering. However, in many design, analysis, and manufacturing tasks, handling the rod-based structures in three dimensions directly is generally challenging. To simplify the tasks, it is usually more desirable to achieve a two-dimensional representation of the rod-based structures via some suitable geometric mappings. In this work, we develop a novel method for computing a low-distortion planar embedding of rod-based structures. Specifically, we identify geometrical constraints that aim to preserve key length and angle quantities of the 3D rod-based structures and prevent the occurrence of overlapping rods in the planar embedding. Experimental results with a variety of rod-based structures are presented to demonstrate the effectiveness of our approach. Moreover, our method can be naturally extended to the design and mapping of hybrid structures consisting of both rods and surface elements. Altogether, our approach paves a new way for the efficient design and fabrication of novel three-dimensional geometric structures for practical applications.

💡 Research Summary

The paper addresses the challenge of representing three‑dimensional rod‑based structures in a two‑dimensional plane while preserving the essential geometric characteristics of the original model. The authors propose a systematic pipeline that begins with an initial planar layout obtained via a Tutte embedding (or a simple orthogonal projection for simple geometries). This initial embedding fixes the dimensionality of the problem, allowing subsequent optimization to be performed entirely in 2‑D space.
The core of the method is a constrained non‑linear optimization that enforces three families of geometric constraints: (1) length preservation, (2) angle preservation at rod joints, and (3) avoidance of rod overlap. Length preservation is encoded as EL(ei)=Li/li−1=0 for each edge, where Li is the planar length and li the original 3‑D length. The authors derive explicit gradients for this term, showing that each edge contributes only four non‑zero entries to the Jacobian, which greatly simplifies the computation. Angle preservation is formulated using cosine values, EA(j)=cos θ2D−cos θ3D=0, for each joint angle. By expressing the cosine as a dot product of two edge vectors and normalizing by their magnitudes, the paper provides detailed partial derivatives with respect to all six coordinates of the three vertices involved, resulting in six non‑zero Jacobian entries per angle constraint.
The most innovative contribution is the “no‑overlap” constraint. Direct pairwise collision checks would be combinatorially expensive, so the authors instead triangulate the initial planar embedding, compute the sum of triangle areas, and enforce equality between this sum and the area enclosed by the boundary vertices: EO=∑Area(Ti)−Area(B)=0. Theorem 1 proves that if this equality holds, the planar configuration cannot contain overlapping rods. The authors also acknowledge that the converse is not true and that the constraint can be overly restrictive for certain configurations; they mitigate this by incorporating an alternating minimization scheme that relaxes the constraint when necessary. Detailed gradient formulas for both the triangle‑area sum and the boundary‑area term are provided.
The optimization is carried out using a quasi‑Newton method (e.g., L‑BFGS), leveraging the analytically derived gradients for rapid convergence. Experiments on a diverse set of structures—including simple frames, complex curved lattices, gridshells, and hybrid rod‑surface models—demonstrate that the method achieves average length errors below 5 % and angle errors below 4°, while guaranteeing zero overlap in the final planar layout. The paper also showcases a 2‑D‑to‑3‑D morphing simulation, confirming that the low‑distortion planar designs can be faithfully reconstructed in three dimensions for fabrication or assembly.
Beyond pure rod networks, the authors extend the framework to hybrid structures that combine one‑dimensional rods with two‑dimensional surface patches, showing that the same constraint formulation can be applied with minor modifications. The discussion highlights limitations such as potential over‑constraining for highly non‑convex shapes and numerical instability in dense, thin rod networks. Future work is suggested in the direction of local overlap constraints, integration with physics‑based simulation for real‑time design feedback, and more sophisticated handling of non‑convex boundaries.
Overall, the paper presents a well‑grounded mathematical approach to low‑distortion planar embedding of rod‑based structures, offering a practical tool for designers and engineers who need to transition between 3‑D models and 2‑D manufacturing or analysis workflows.