ArtiCAD: Articulated CAD Assembly Design via Multi-Agent Code Generation

2026-04-13

AI Summary

Automatically generating editable, kinematically functional parametric CAD assemblies from high-level text or image inputs remains a significant challenge. This work proposes the first training-free multi-agent system that coordinates four specialized agents—design, generation, assembly, and review—to explicitly predict assembly relationships during the design phase via a connector-based mechanism. The framework incorporates multi-stage verification, cross-phase rollback, and a self-evolving experience repository to ensure output fidelity. By circumventing the spatial reasoning limitations of large language models, the approach supports code generation, joint definition, and experience reuse. Validated on ArtiCAD-Bench, CADPrompt, and ACD datasets, the method demonstrates effectiveness in conceptual design and physical prototyping, and can export URDF files for embodied AI training.

Technology Category

• Cognitive Modeling & Cognitive Systems: Agent Architectures
• Multiagent Systems: Adversarial Agents
• Intelligent Robots: Embodied AI

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: Agentic search
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Parametric Computer-Aided Design (CAD) of articulated assemblies is essential for product development, yet generating these multi-part, movable models from high-level descriptions remains unexplored. To address this, we propose ArtiCAD, the first training-free multi-agent system capable of generating editable, articulated CAD assemblies directly from text or images. Our system divides this complex task among four specialized agents: Design, Generation, Assembly, and Review. One of our key insights is to predict assembly relationships during the initial design stage rather than the assembly stage. By utilizing a Connector that explicitly defines attachment points and joint parameters, ArtiCAD determines these relationships before geometry generation, effectively bypassing the limited spatial reasoning capabilities of current LLMs and VLMs. To further ensure high-quality outputs, we introduce validation steps in the generation and assembly stages, accompanied by a cross-stage rollback mechanism that accurately isolates and corrects design- and code-level errors. Additionally, a self-evolving experience store accumulates design knowledge to continuously improve performance on future tasks. Extensive evaluations on three datasets (ArtiCAD-Bench, CADPrompt, and ACD) validate the effectiveness of our approach. We further demonstrate the applicability of ArtiCAD in requirement-driven conceptual design, physical prototyping, and the generation of embodied AI training assets through URDF export.

Problem

• Articulated CAD
• Parametric Design
• Assembly Generation
• Multi-part Modeling
• Movable Assemblies

Innovation

• multi-agent system
• articulated CAD assembly
• connector-based assembly
• cross-stage rollback
• self-evolving experience store