DYNAMO: Dependency-Aware Deep Learning Framework for Articulated Assembly Motion Prediction

2025-09-15

AI Summary

Inferring coupled motion in articulated mechanical assemblies (e.g., gear systems) from static CAD models remains challenging, as geometric cues alone are insufficient for reliable kinematic reasoning without explicit joint annotations. Method: This paper introduces MechBench—the first benchmark dataset specifically designed for gear assemblies—and proposes DYNAMO, an end-to-end learnable framework that takes segmented CAD point clouds as input. DYNAMO employs perception-driven neural networks to jointly model geometric contact and power transmission relationships among parts, integrating contact inference with motion propagation to predict SE(3) motion trajectories without requiring joint labels. Contribution/Results: Evaluated on diverse gear configurations, DYNAMO significantly outperforms strong baselines, achieving high-accuracy, temporally consistent, part-level motion prediction. Its robust performance demonstrates both effectiveness and strong generalization across unseen gear topologies and configurations.

Technology Category

• Computer Vision: Motion & Tracking
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning
• Intelligent Robots: Motion and Path Planning

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Understanding the motion of articulated mechanical assemblies from static geometry remains a core challenge in 3D perception and design automation. Prior work on everyday articulated objects such as doors and laptops typically assumes simplified kinematic structures or relies on joint annotations. However, in mechanical assemblies like gears, motion arises from geometric coupling, through meshing teeth or aligned axes, making it difficult for existing methods to reason about relational motion from geometry alone. To address this gap, we introduce MechBench, a benchmark dataset of 693 diverse synthetic gear assemblies with part-wise ground-truth motion trajectories. MechBench provides a structured setting to study coupled motion, where part dynamics are induced by contact and transmission rather than predefined joints. Building on this, we propose DYNAMO, a dependency-aware neural model that predicts per-part SE(3) motion trajectories directly from segmented CAD point clouds. Experiments show that DYNAMO outperforms strong baselines, achieving accurate and temporally consistent predictions across varied gear configurations. Together, MechBench and DYNAMO establish a novel systematic framework for data-driven learning of coupled mechanical motion in CAD assemblies.

Problem

• Predicting motion trajectories from static geometry
• Understanding coupled motion in gear assemblies
• Overcoming reliance on predefined joint annotations

Innovation

• Dependency-aware neural model for motion prediction
• Predicts SE(3) trajectories from CAD point clouds
• Handles coupled motion through geometric contact transmission