Bai Yunpeng, Zhao Min, Li Yuanjun, Zhang Haonan, Zhao Chenjie, Liu Bingjun, Tian Xiaoquan, Chen Dengkai
School of Mechatronic Engineering, Northwestern Polytechnical University, Xi'an, 710072, China.
Ningbo Institute of Northwestern Polytechnical University, Ningbo, 315103, China.
Sci Rep. 2025 Sep 1;15(1):32223. doi: 10.1038/s41598-025-17164-1.
This study aims to develop a human-machine co-creation framework for automobile seat conceptual design, leveraging an improved Deep Convolutional Generative Adversarial Network (ResNet-DCGAN) to lower design barriers for non-professionals and enhance cross-disciplinary innovation. By constructing a dataset of automobile seat images and implementing generative design strategies across three key stages, this research seeks to demonstrate the feasibility of AI-driven creativity augmentation in product design. The cooperation of human-machine co-creation can stimulate the innovative thinking of participants from different industries, reduce the design difficulty, arouse participants' enthusiasm, and provide abundant creativity in designing automobile seats. At each stage of the automobile seat design, methodological strategies are proposed to promote the successful implementation of the design. Specifically, by creating a data set of automobile seats and using the deep convolutional generative adversarial network improved by the ResNet residual module (ResNet-DCGAN) to stimulate creative inspiration, we let the computer take the preliminary sketches of the participants as input to generate design schemes that are both innovative and meet the aesthetic requirements of the public, providing a theoretical basis for human-machine collaborative innovation research.
本研究旨在开发一种用于汽车座椅概念设计的人机共创框架，利用改进的深度卷积生成对抗网络（ResNet-DCGAN）降低非专业人士的设计门槛，增强跨学科创新。通过构建汽车座椅图像数据集并在三个关键阶段实施生成式设计策略，本研究旨在证明人工智能驱动的产品设计创意增强的可行性。人机共创的合作可以激发不同行业参与者的创新思维，降低设计难度，激发参与者的热情，并在汽车座椅设计中提供丰富的创意。在汽车座椅设计的每个阶段，都提出了方法策略以促进设计的成功实施。具体而言，通过创建汽车座椅数据集并使用由ResNet残差模块改进的深度卷积生成对抗网络（ResNet-DCGAN）来激发创意灵感，我们将参与者的初步草图作为输入，让计算机生成既创新又符合大众审美要求的设计方案，为 人机协同创新研究提供理论依据。
