Physics-in-the-Loop: A Hybrid Agentic Architecture for Validated CAD Engineering Design

Berger, Elias; Usama, Muhammad; Mehlstäubl, Jan; Saske, Bernhard; Paetzold-Byhain, Kristin

doi:10.24963/ijcai.2026/714

Physics-in-the-Loop: A Hybrid Agentic Architecture for Validated CAD Engineering Design

Elias Berger, Muhammad Usama, Jan Mehlstäubl, Bernhard Saske, Kristin Paetzold-Byhain

Download BibTeX Cite

Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence. AI4Tech: AI Enabling Technologies. Pages 6417-6425.

https://doi.org/10.24963/ijcai.2026/714

Keywords

Advanced AI4Tech: Generative and LLMs-driven AI4Tech
Domain-specific AI4Tech: AI4Manufacturing
AI4T: Domain-specific AI4Tech

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 4.2x 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.

Elias Berger, Muhammad Usama, Jan Mehlstäubl, Bernhard Saske, and Kristin Paetzold-Byhain. 2026. Physics-in-the-Loop: A Hybrid Agentic Architecture for Validated CAD Engineering Design. In Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence, IJCAI-26. IJCAI, 6417–6425. https://doi.org/10.24963/ijcai.2026/714

@inproceedings{ijcai2026p714,
  title     = {Physics-in-the-Loop: A Hybrid Agentic Architecture for Validated CAD Engineering Design},
  author    = {Berger, Elias and Usama, Muhammad and Mehlstäubl, Jan and Saske, Bernhard and Paetzold-Byhain, Kristin},
  booktitle = {Proceedings of the Thirty-Fifth International Joint Conference on
               Artificial Intelligence, {IJCAI-26}},
  publisher = {International Joint Conferences on Artificial Intelligence Organization},
  editor    = {Diego Calvanese},
  pages     = {6417--6425},
  year      = {2026},
  month     = {8},
  note      = {AI4Tech: AI Enabling Technologies},
  doi       = {10.24963/ijcai.2026/714},
  url       = {https://doi.org/10.24963/ijcai.2026/714},
}

TY  - CPAPER
TI  - Physics-in-the-Loop: A Hybrid Agentic Architecture for Validated CAD Engineering Design
AU  - Berger, Elias
AU  - Usama, Muhammad
AU  - Mehlstäubl, Jan
AU  - Saske, Bernhard
AU  - Paetzold-Byhain, Kristin
T2  - Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence, IJCAI-26
PB  - International Joint Conferences on Artificial Intelligence Organization
PY  - 2026
SP  - 6417
EP  - 6425
DO  - 10.24963/ijcai.2026/714
UR  - https://doi.org/10.24963/ijcai.2026/714
ER  -