SBIR Phase I: AI Robotics-driven Material Discovery Platform

这个小型企业创新研究(SBIR)第一阶段项目的更广泛影响/商业潜力是利用人工智能(AI)机器人驱动的材料开发平台加快新的高性能电池材料的开发。该平台使用机器学习和机器人高通量自动化来加速有效的实验规划并将错误降至最低。它可能会对优质电池材料的商业化产生重大的积极影响(预计到2025年将达到140亿美元的市场)，以支持电动汽车和其他可持续交通的增长。这个小型企业创新研究(SBIR)一期项目旨在建立一个以闭环机器学习和机器人高通量自动化为特色的材料开发平台，并开发用于锂电池的高性能聚合物电解液产品。该平台可能会改变材料创新的方式，并加速发现电解液和其他电池材料。该平台的工作流程迭代如下：(1)初始电解液知识库收集；(2)使用知识库进行机器学习模型训练；(3)通过模型进行新的电解液配方；(4)通过高通量设备进行并行实验验证；以及(5)知识库更新。第一阶段将帮助(1)在用于电解液开发的机器人系统上建立关键的电化学和机械模块，(2)从可行性、灵活性和优化多目标函数的能力方面改进机器学习模型，(3)开发聚合物电解液配方，以改善其三个主要性能，包括离子导电性、电压稳定性和机械模数。预计该平台将实现高生产率和有效性，显著改善电解液性能，并确定符合商业化系统要求的电解液。该奖项反映了NSF的法定使命，并通过使用基金会的智力优势和更广泛的影响审查标准进行评估，被认为值得支持。

The broader impact/commercial potential of this Small Business Innovation Research (SBIR) Phase I project is to accelerate the development of new high-performance battery materials with an Artificial Intelligence (AI) robotics-driven material development platform. The platform uses machine learning and robotic high-throughput automation to accelerate effective experiment planning and minimize errors. It will potentially have a substantial positive impact on the commercialization of superior battery materials (projected to be a $14 B market by 2025), to support growth of electric vehicles and other sustainable transportation. This Small Business Innovation Research (SBIR) Phase I project aims to build a material development platform featuring a closed-loop machine learning and robotic high-throughput automation, and to develop a high-performance polymer electrolyte product for lithium batteries. The platform can potentially change how material innovation is performed and enable accelerated discovery of electrolytes and other battery materials. The platform’s workflow iterates the following: (1) initial electrolyte knowledge base collection; (2) machine-learning model training using the knowledge base; (3) new electrolyte prescription by the model; (4) parallelized experimental validation via high-throughput equipment; and (5) knowledge base updates. Phase I will help to (1) build key electrochemical and mechanical modules on the robotic system for electrolyte development, (2) improve machine learning models in terms of feasibility, flexibility, and the capability of optimizing multiple objective functions, and (3) develop the polymer electrolyte formulation in order to improve its three primary properties, including ionic conductivity, voltage stability, and mechanical modulus. It is anticipated that the platform will achieve high productivity and effectiveness, significantly improve electrolyte properties, and identify an electrolyte that meets commercialization system requirements.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.