Creative synthesis of kinematic mechanisms

2025-09-30

AI Summary

This work addresses the inverse design problem of planar linkage mechanism synthesis by reformulating the trajectory–structure mapping as a cross-domain image generation task. Methodologically, we construct an RGB-image-based dataset of mechanism–trajectory pairs and propose a shared-latent-space variational autoencoder framework. Crucially, we introduce a novel color-gradient encoding scheme to embed trajectory velocity information, enabling joint conditional control over both geometric shape and kinematic (velocity) properties. The approach supports end-to-end inverse design of diverse mechanisms—including revolute-joint linkages, sliding mechanisms, and multi-loop topologies with cams, gears, and prismatic joints. Experiments on four-bar, slider-crank, and multi-loop mechanism datasets demonstrate high-fidelity synthesis of feasible mechanism configurations for unseen trajectory curves, validating the effectiveness and generalization capability of image-driven mechanical innovation design.

Technology Category

• Computer Vision: Computational Photography, Image & Video Synthesis
• Machine Learning: Imitation Learning & Inverse Reinforcement Learning
• Intelligent Robots: Motion and Path Planning

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• Economics, Online Markets and Human Computation: Economic ramifications for generative AI infrastructure and applications

Abstract

In this paper, we formulate the problem of kinematic synthesis for planar linkages as a cross-domain image generation task. We develop a planar linkages dataset using RGB image representations, covering a range of mechanisms: from simple types such as crank-rocker and crank-slider to more complex eight-bar linkages like Jansen's mechanism. A shared-latent variational autoencoder (VAE) is employed to explore the potential of image generative models for synthesizing unseen motion curves and simulating novel kinematics. By encoding the drawing speed of trajectory points as color gradients, the same architecture also supports kinematic synthesis conditioned on both trajectory shape and velocity profiles. We validate our method on three datasets of increasing complexity: a standard four-bar linkage set, a mixed set of four-bar and crank-slider mechanisms, and a complex set including multi-loop mechanisms. Preliminary results demonstrate the effectiveness of image-based representations for generative mechanical design, showing that mechanisms with revolute and prismatic joints, and potentially cams and gears, can be represented and synthesized within a unified image generation framework.

Problem

• Formulating kinematic synthesis as cross-domain image generation task
• Developing image generative models for synthesizing unseen motion curves
• Supporting kinematic synthesis conditioned on trajectory shape and velocity

Innovation

• Uses shared-latent VAE for cross-domain kinematic synthesis
• Encodes trajectory speed as color gradients in images
• Unified image framework represents various joint mechanisms