Reinforcement learning for inverse structural design and rapid laser cutting of kirigami prototypes

Milad Yazdani, Shahriar Shalileh, Dena Shahriari

arXiv:2605.08098 · 2026-05-12 공개 · arXiv · PDF

reinforcement-learning flow-matching grpo optimal-transport inverse-design deployment-simulation kirigami silhouette-matching

Abstract

Kirigami is an increasingly useful fabrication method to produce shape-programmable metamaterial structures. However, inverse design remains difficult because deployment is nonlinear, and feasible cut layouts must satisfy discrete compatibility rules, avoid overlap, and map one target shape to valid designs. We present RL-Kirigami, an inverse design framework that combines optimal-transport conditional flow matching (OT-CFM) with reinforcement learning to generate compatible ratio fields for compact reconfigurable parallelogram quad kirigami. A marching decoder enforces global geometric compatibility, and Group Relative Policy Optimization (GRPO) aligns the generator with nondifferentiable rewards for silhouette matching, feasibility, and ratio-field regularity. Across procedurally generated target shape instances, a single sample from the pretrained OT-CFM prior reached $94.2%$ sIoU and outperformed solver baselines while reducing forward simulator evaluations from hundreds to 1. GRPO improved accuracy to $94.91%$ sIoU and, with regularity included, reduced $\mathrm{TV}(\mathbf{x})$ from 0.95 to 0.81 while maintaining $94.83%$ sIoU. Generated layouts were exported to DXF and laser-cut in $50~\mu\mathrm{m}$ polymeric sheets to produce deployable prototypes in $8.0 \pm 1.0$ minutes per part. These results support a manufacturing-aware inverse design workflow for deployable kirigami metamaterials under hard geometric feasibility constraints.

한국어 요약

📋 한 줄 요약

**[메타재료 / 강화학습 역설계]** 키리가미 메타재료의 호환성 제약을 만족하는 reconfigurable parallelogram quad 키리가미를 OT-CFM과 RL로 역설계하고 실제 레이저 컷팅까지 검증한 RL-Kirigami 프레임워크 제안.

🎯 핵심 기여도

💡 핵심 아이디어

미분 불가능한 보상(실루엣 일치·feasibility·정규성)을 RL로 처리하고, 호환 가능한 ratio field를 OT-CFM 사전으로 빠르게 샘플링한 뒤 RL로 세밀 정렬하면 시뮬레이터 호출 수를 수백 회에서 1회로 줄일 수 있다는 통찰.

🔬 기술적 접근법

📊 주요 결과

💭 의의 및 한계

**의의**: 시뮬레이션·생성 모델·RL·제조를 연결하는 manufacturing-aware 역설계 워크플로의 모범 사례. **한계**: parallelogram quad 키리가미라는 특정 클래스에 집중, 다른 키리가미 패턴 클래스로의 확장은 추가 연구 필요.

🚀 실용적 활용