🤖 AI Summary
This study addresses the reliance on manual labor, fragmented workflows, and lack of collaborative oversight in converting CAD models into assembly documentation. To overcome these challenges, this work proposes an automated assembly instruction generation framework grounded in large language models. The framework employs a five-layer architecture that leverages STEP parsing and a ProductGraph intermediate representation to achieve structured information extraction and deterministic topological sorting. By integrating constrained contextual generation with visual augmentation, it realizes an end-to-end pipeline, while incorporating rule-based verification and human-in-the-loop mechanisms to ensure engineering accuracy. Experimental results validate the effectiveness of this system, which represents the first unified closed-loop approach to seamlessly integrate CAD interpretation, sequence planning, instruction writing, and human supervision.
📝 Abstract
Assembly documentation is a downstream manufacturing artifact that is still usually authored by interpreting CAD models by hand. Structured product data and large language models are both available, yet studies of CAD interpretation, assembly sequence planning, instruction writing, and human oversight have largely proceeded separately. This paper formulates CAD-grounded assembly instruction generation: the production of natural-language assembly procedures constrained by structured engineering information extracted from CAD models. The proposed framework maps a STEP assembly to a typed ProductGraph intermediate representation, derives a precedence order by deterministic topological sorting, realizes each step as language conditioned only on selected graph context, attaches per-step visual documentation, and applies rule-based and model-assisted checks. PDF export remains disabled until a human reviewer resolves every quality flag. The case study establishes endto-end feasibility on a built-in six-part reference assembly: the pipeline preserves a reported assembly order and carries quantity, material, and torque into an exported manual page. Generalization and geometric validation remain open empirical questions. The contribution is an architecture that separates engineering state, deterministic reasoning, grounded language realization, verification, and human release.