Transformer-Based Interfaces for Mechanical Assembly Design: A Gear Train Case Study

2025-04-11

AI Summary

This study addresses the challenge of deep integration between generative AI and mechanical assembly design tools, specifically for gear transmission mechanism design. We propose a dual-mode Transformer interaction paradigm: the *Explore* mode employs probabilistic sampling to generate and evaluate multiple design alternatives, enabling problem redefinition and creative exploration; the *Copilot* mode leverages context-aware autoregressive sequence prediction to support controllable, iterative optimization. The model architecture is customized for mechanical assembly semantics and integrated with a human-centered hybrid workflow and interactive UI. A user study demonstrates that *Explore* significantly improves exploratory efficiency and problem reframing capability, while *Copilot* enhances engineers’ control and engagement depth. Together, the two modes effectively support complex, creative design tasks—marking the first systematic incorporation of sequence modeling and probabilistic generation into an end-to-end engineering design loop.

Technology Category

• Humans and AI: Game Design — Procedural Content Generation & Storytelling
• Natural Language Processing: Generation
• Machine Learning: Deep Generative Models & Autoencoders

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Economics, Online Markets and Human Computation: Uses of LLMs and GenAI for marketplace design, bidding, and strategic interactions

Abstract

Generative artificial intelligence (AI), particularly transformer-based models, presents new opportunities for automating and augmenting engineering design workflows. However, effectively integrating these models into interactive tools requires careful interface design that leverages their unique capabilities. This paper introduces a transformer model tailored for gear train assembly design, paired with two novel interaction modes: Explore and Copilot. Explore Mode uses probabilistic sampling to generate and evaluate diverse design alternatives, while Copilot Mode utilizes autoregressive prediction to support iterative, context-aware refinement. These modes emphasize key transformer properties (sequence-based generation and probabilistic exploration) to facilitate intuitive and efficient human-AI collaboration. Through a case study, we demonstrate how well-designed interfaces can enhance engineers' ability to balance automation with domain expertise. A user study shows that Explore Mode supports rapid exploration and problem redefinition, while Copilot Mode provides greater control and fosters deeper engagement. Our results suggest that hybrid workflows combining both modes can effectively support complex, creative engineering design processes.

Problem

• Develop transformer-based interfaces for gear train assembly design
• Integrate Explore and Copilot modes for human-AI collaboration
• Balance automation with domain expertise in engineering workflows

Innovation

• Transformer model tailored for gear train design
• Explore Mode generates diverse design alternatives
• Copilot Mode enables iterative context-aware refinement