Non-Iterative Inverse Design for 4D Printing via Fabrication-Ready Discrete Material Programming

Description

This deposit supports Virtual and Physical Prototyping research on non-iterative inverse design for 4D printing using fabrication-ready discrete material programming. It provides (i) datasets of discrete printing patterns with semantic labels (including void) and optional FEA-derived midplane fields for correlation studies, and (ii) trained segmentation model weights (SegFormer-style, Hugging Face format) used to map coordinate-based inputs to discrete material layouts. Together with the companion code repository, these resources enable reproducibility of the reported pipeline from pattern definition and perturbation through learning-based inference to fabrication-oriented analysis.

Categories

Licence