Tool-Augmented Agent for Closed-loop Optimization,Simulation,and Modeling Orchestration

Liyuan Deng, Shujian Deng, Yongkang Chen, Yongkang Dai, Zhihang Zhong, Linyang Li, Xiao Sun, Yilei Shi, Huaxi Huang

arXiv:2605.20190 · 2026-05-22 공개 · arXiv · PDF

reward-shaping llm-orchestration constraint-driven industrial-ai simulation-modeling parametric-geometry tool-augmented-rl design-simulation

Abstract

Iterative industrial design-simulation optimization is bottlenecked by the CAD-CAE semantic gap: translating simulation feedback into valid geometric edits under diverse, coupled constraints. To fill this gap, we propose COSMO-Agent (Closed-loop Optimization, Simulation, and Modeling Orchestration), a tool-augmented reinforcement learning (RL) framework that teaches LLMs to complete the closed-loop CAD-CAE process. Specifically, we cast CAD generation, CAE solving, result parsing, and geometry revision as an interactive RL environment, where an LLM learns to orchestrate external tools and revise parametric geometries until constraints are satisfied. To make this learning stable and industrially usable, we design a multi-constraint reward that jointly encourages feasibility, toolchain robustness, and structured output validity. In addition, we contribute an industry-aligned dataset that covers 25 component categories with executable CAD-CAE tasks to support realistic training and evaluation. Experiments show that COSMO-Agent training substantially improves small open-source LLMs for constraint-driven design, exceeding large open-source and strong closed-source models in feasibility, efficiency, and stability.

한국어 요약

한 줄 요약

**[Industrial CAD-CAE Agent / RL]** COSMO-Agent가 CAD 생성·CAE 해석·결과 파싱·기하 수정을 RL 환경으로 다중 제약 reward로 학습, 25 component category·industrial-aligned 데이터셋에서 소형 오픈소스 LLM이 대형·강한 폐쇄형 모델 능가.

핵심 기여도

핵심 아이디어

CAD-CAE closed-loop 자동화의 본질은 LLM이 시뮬레이션 피드백을 valid 기하 수정으로 번역하는 능력이며, tool-augmented RL의 interactive 환경에서 feasibility·robustness·validity를 jointly 권장하는 reward로 학습하면 소형 오픈소스 LLM도 산업 수준에 도달한다.

기술적 접근법

주요 결과

의의 및 한계

**의의**: CAD-CAE semantic gap 해소의 LLM 기반 첫 RL 통합, 다중 제약 reward로 산업 신뢰성 보장. **한계**: 25 component category 외 generalization·다른 엔지니어링 도메인 확장은 추가 검증 필요.

실용적 활용