SBIR Phase I: Methods for Embedding User Data into 3D Generative AI Computer-aided-Design Models

SBIR 第一阶段：将用户数据嵌入 3D 生成式 AI 计算机辅助设计模型的方法

• 批准号: 2335491
• 负责人: Calvin Smith
• 金额: $ 27.5万
• 依托单位: ADA TECH LLC
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2024
• 资助国家: 美国
• 起止时间: 2024-03-01 至 2025-02-28
• 项目状态: 未结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2335491&HistoricalAwards=false
• 关键词: SBIR; Phase; Methods; Embedding; User

The broader/commercial impact of this Small Business Innovation Research (SBIR) Phase I project is the development of a novel Artificial-Intelligence-powered generative design solution that is able to address the needs of industrial and consumer-product manufacturers by exploiting the abundance of data (social media, usage, telemetry) currently available. The proposed framework will create new opportunities for American design and manufacturing firms to better align their products with rapidly evolving consumer needs while reducing the product development challenges that currently exist. It also enables faster cycle times for product development and the near-real-time inclusion of consumer sentiment into product design. The proposed computational methods will translate consumers’ digital insight into new ways to increase the quality of the design concepts and the diversity of consumer perspectives incorporated into AI-generated design concepts, thereby enhancing the designers’ ability to innovate socially aware, consumer-centric products. This project will foster the design of novel, effective, and efficient design models, augment designers’ creativity, promote designer-AI co-creation and bias mitigation, and bridge the gap between consumer-needs discovery, Design for Excellence (DFX) engineering, and social impact. This has ramifications for nearly every industry and application.This Small Business Innovation Research (SBIR) Phase I project will enable a generational leap in three-dimensional generative design capabilities by integrating qualitative and quantitative information into generative AI models for the efficient production of novel designs. The primary objective is to develop a testable demonstrator for fusing consumer data, data from the Internet of Things (IoT), and Design for Excellence (DFX) engineering specifications into 3D geometric data. The Phase I project will focus on exploring new methods for natural language processing, generative modeling, and data fusion models to integrate consumer data and technical requirements with IoT-based telemetric data, drawing inferences for product design, and building novel semi-supervised models to inject these data inputs into 3D CAD generative models. The project will determine how to directly connect consumer needs with functional performance and study the real-world effectiveness and efficiency gains from the generalizability of 3D generative design. The project will address several challenges of current generative design solutions, including the translation of qualitative and quantitative metadata into concepts, the control and iteration of automatically generated designs, and their seamless integration into manufacturing processes and workflows.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

这项小型企业创新研究（SBIR）I期项目的更广泛/商业影响是开发一种新型的人工智力供电的通用设计解决方案，该解决方案能够通过利用数据的抽象（社交媒体，用法，远程信息组）来解决工业和消费者产品制造商的需求。拟议的框架将为美国设计和制造公司创造新的机会，以使其产品与迅速发展的消费者需求更好地对齐，同时减少当前存在的产品开发挑战。它还可以使产品开发以及将消费者情感纳入产品设计的速度更快的周期时间。提出的计算方法将把消费者的数字洞察力转化为新的方法，以提高设计概念的质量以及纳入AI生成的设计概念的消费者观点的多样性，从而增强了设计师创新的社会意识，以消费者为中心的产品。该项目将促进新颖，有效且高效的设计模型的设计，增强设计师的创意，促进设计师-AI共同创造和缓解偏见，并弥合消费者需求发现，卓越设计（DFX）工程设计和社会影响之间的差距。这几乎对每个行业和应用都有后果。这项小型企业创新研究（SBIR）I阶段项目将通过将定性和定量信息集成到有效生产新颖设计的通用AI模型中，从而使三维通用设计能力成为世代相传的飞跃。主要目的是开发一个可测试的演示器，以融合消费者数据，物联网（IoT）的数据以及卓越设计（DFX）工程规范，以将其融合到3D几何数据中。 I阶段项目将着重于探索自然语言处理，通用建模和数据融合模型的新方法，以将消费者数据和技术需求与基于IoT的远程测试数据集成，绘制产品设计的推断以及构建新颖的半私人模型，以将这些数据输入注入3D CAD CAD CAD Generic模型中。该项目将确定如何将消费者需求与功能性能联系起来，并研究3D通用设计的普遍性中现实世界的有效性和效率提高。 The project will address Several challenges of current generic design solutions, including the translation of qualitative and quantitative metadata into concepts, the control and iteration of automatically generated designs, and their seamless integration into manufacturing processes and workflows.This award reflects NSF's statutory mission and has been deemed precious of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.