Automated Assembly Instruction Generation from CAD Models Using Grounded Large Language Models: A Human-in-the-Loop Framework
The framework generates natural-language assembly instructions from CAD models by mapping them to a structured ProductGraph. It uses deterministic topological sorting for ordering, grounds language generation in graph context, and incorporates rule-based checks and human verification before releasing the final documentation. This means engineering data dictates the structure, while language provides the descriptive steps.