Saved in:
Bibliographic Details
Main Author: Wilke, Daniel N.
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2604.06788
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917401033965568
author Wilke, Daniel N.
author_facet Wilke, Daniel N.
contents We present a solver-agnostic framework in which coordinated large language model (LLM) agents autonomously execute the complete computational mechanics workflow, from perceptual data of an engineering component through geometry extraction, material inference, discretisation, solver execution, uncertainty quantification, and code-compliant assessment, to an engineering report with actionable recommendations. Agents are formalised as conditioned operators on a shared context space with quality gates that introduce conditional iteration between pipeline layers. We introduce a mathematical framework for extracting engineering information from perceptual data under uncertainty using interval bounds, probability densities, and fuzzy membership functions, and introduce task-dependent conservatism to resolve the ambiguity of what `conservative' means when different limit states are governed by opposing parameter trends. The framework is demonstrated through a finite element analysis pipeline applied to a photograph of a steel L-bracket, producing a 171,504-node tetrahedral mesh, seven analyses across three boundary condition hypotheses, and a code-compliant assessment revealing structural failure with a quantified redesign. All results are presented as generated in the first autonomous iteration without manual correction, reinforcing that a professional engineer must review and sign off on any such analysis.
format Preprint
id arxiv_https___arxiv_org_abs_2604_06788
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle From Perception to Autonomous Computational Modeling: A Multi-Agent Approach
Wilke, Daniel N.
Computational Engineering, Finance, and Science
Computation and Language
Multiagent Systems
65N30, 68T42, 74S05
I.2.11; I.2; J.2
We present a solver-agnostic framework in which coordinated large language model (LLM) agents autonomously execute the complete computational mechanics workflow, from perceptual data of an engineering component through geometry extraction, material inference, discretisation, solver execution, uncertainty quantification, and code-compliant assessment, to an engineering report with actionable recommendations. Agents are formalised as conditioned operators on a shared context space with quality gates that introduce conditional iteration between pipeline layers. We introduce a mathematical framework for extracting engineering information from perceptual data under uncertainty using interval bounds, probability densities, and fuzzy membership functions, and introduce task-dependent conservatism to resolve the ambiguity of what `conservative' means when different limit states are governed by opposing parameter trends. The framework is demonstrated through a finite element analysis pipeline applied to a photograph of a steel L-bracket, producing a 171,504-node tetrahedral mesh, seven analyses across three boundary condition hypotheses, and a code-compliant assessment revealing structural failure with a quantified redesign. All results are presented as generated in the first autonomous iteration without manual correction, reinforcing that a professional engineer must review and sign off on any such analysis.
title From Perception to Autonomous Computational Modeling: A Multi-Agent Approach
topic Computational Engineering, Finance, and Science
Computation and Language
Multiagent Systems
65N30, 68T42, 74S05
I.2.11; I.2; J.2
url https://arxiv.org/abs/2604.06788