Physics-in-the-Loop: A Hybrid Agentic Architecture for Validated CAD Engineering Design
Abstract
Large Language Models (LLMs) can generate Computer-Aided Design (CAD), yet lack physical comprehension required for reliable engineering design. Instead of attempting to implicitly learn physical laws from data, we propose a Hybrid Agentic-Physical Architecture that embeds validated knowledge-based engineering tools directly into the decision making loop of autonomous AI agents. In this framework, engineering design is formulated as a closed-loop, sequential decision making process guided by explicit physical verification. Based on a load case, dedicated agents iteratively plan, generate, evaluate, and revise engineering designs using knowledge-based tools as a feedback signal. We introduce a benchmark dataset and metrics for assessing functional validity in generative CAD. Our system generates more complex and physically verified designs, with a 4.2 increase in structural complexity and improving compile rate by 3.5% compared to similar agentic methods. The codebase, prompts and dataset will be made publicly available to support reproducibility and future research.
Cite
@article{arxiv.2605.19717,
title = {Physics-in-the-Loop: A Hybrid Agentic Architecture for Validated CAD Engineering Design},
author = {Elias Berger and Muhammad Usama and Jan Mehlstäubl and Bernhard Saske and Kristin Paetzold-Byhain},
journal= {arXiv preprint arXiv:2605.19717},
year = {2026}
}
Comments
Accepted in IJCAI-ECAI 2026 (Special Track on AI4Tech)