Session: DAC-04-03: Data-Driven Design

Paper Number: 114464

114464 - Deep Generative Model-Based Synthesis of Four-Bar Linkage Mechanisms Considering Both Kinematic and Dynamic Conditions 

Mechanisms are essential components designed to perform specific tasks in various mechanical systems. However, designing a mechanism that satisfies both kinematic and dynamic requirements is a challenging task. The kinematic requirements of a mechanism include its workspace, which is the region in which the mechanism can operate, where the maximum scalar displacement that the mechanism can be derived from this region. On the other hand, the dynamic requirements of a mechanism include its torque transmission, which refers to the ability of the mechanism to transfer power and rotational force effectively. In this paper, we propose a deep generative model that can generate multiple crank-rocker four-bar linkage mechanism samples that satisfy both the kinematic and dynamic requirements aforementioned. The proposed model is based on a conditional generative adversarial network (cGAN) with some modifications for mechanism synthesis, which is trained to learn the relationship between the requirements of a mechanism with respect to linkage lengths, and generates multiple mechanism samples that satisfy given requirements. The results demonstrate that our proposed method can successfully generate multiple mechanisms that satisfy specific requirements. Our approach has several advantages over traditional design methods. It enables designers to explore a larger design space and efficiently generate multiple diverse and feasible designs. Also, the proposed method considers both the kinematic and dynamic requirements, which can lead to more efficient and effective mechanisms for real-world use, making it a promising tool for linkage mechanism design.

Presenting Author: Sumin Lee Korea Advanced Institute of Science and Technology

Presenting Author Biography: Sumin Lee received his B.S. degree in Mechanical Engineering from Pusan National University (PNU), Busan, Republic of Korea, in 2020 and M.S. degree in Mechanical Engineering from Korea Advanced Institute of Science and Technology (KAIST), Deajeon, Republic of Korea, in 2022. He is currently studying Ph.D. in Mechanical Engineering at KAIST. His research interests include robotics, mechanism design, and AI-based mechanical design.

Authors:

Sumin Lee Korea Advanced Institute of Science and Technology
Jihoon Kim Korea Advanced Institute of Science and Technology
Namwoo Kang Korea Advanced Institute of Science and Technology