From Perception to Autonomous Computational Modeling: A Multi-Agent Approach

TL;DR

This paper introduces a multi-agent framework enabling autonomous execution of engineering workflows from perceptual data to actionable reports, integrating LLMs with formalized processes and uncertainty management.

Contribution

It presents a novel, solver-agnostic multi-agent system that automates complex engineering analysis workflows with formalized decision gates and uncertainty quantification.

Findings

Successfully analyzed a steel L-bracket using autonomous LLM agents.

Generated a detailed finite element mesh and multiple boundary condition analyses.

Produced a code-compliant structural assessment with quantified failure risk.

Abstract

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.