IUCRC Planning Grant University of Michigan Dearborn: Center for Pervasive Personalized Intelligence Center

物联网（Internet of Things, IoT）是指计算设备（如手机、家用电器、汽车等）通过互联网相互连接的生态系统。普及个性化智能（PPI）中心的目标是使下一代物联网系统在不影响安全和隐私的情况下更加主动和个性化，从而改善决策，提高效率并实现新型计算应用。该提案将与工业界以及科罗拉多大学博尔德分校（CU）和俄勒冈州立大学（OSU）的两个PPI站点合作，研究在密歇根大学迪尔伯恩分校（UM-Dearborn）建立一个新的PPI站点的可行性。密歇根大学迪尔伯恩分校的研究人员将在三个重点下进行基础研究：智能交通、智能能源和软件工程智能。首先，在智能交通的推动下，研究人员将开发新的算法和技术，通过人工智能（即智能个性化）实现联网和自动化车辆的个性化和情境感知机动。其次，对于智能能源推力，研究人员将创造新技术，为电力产消者提供新颖的人工智能辅助个性化需求响应技术。最后，在软件工程智能的推动下，研究人员将探索帮助软件开发人员构建优化的PPI应用程序的方法，这些应用程序可以在网络物理系统中高效、安全地部署和发展。物联网的进步影响着日常生活的许多方面。PPI中心探索人工智能、计算系统、可编程性、能源、可视化和计算机安全与隐私等领域的研究重点，这些领域广泛适用于物联网系统以外的各种其他科学领域，如环境科学、农业、生物、交通运输和机器人技术。研究成果将通过出版物、向工业界和学术界发表研究报告、开源代码和工业合作等方式传播。PPI中心致力于扩大计算机领域的参与，例如，通过与密歇根大学迪尔伯恩分校的项目合作，为工业工程师开设暑期学校，并推广K-12课程。为了提高对未被充分代表的人口的保留，拟议中的中心将与技术协会合作，展示有助于人类的PPI应用。PPI中心将维护一个可公开访问的项目存储库，该存储库将存储开源代码、公开可用的数据集、论文和其他研究成果，这些成果可以在ppicenter.org上向公众传播。该奖项反映了NSF的法定使命，并通过使用基金会的智力价值和更广泛的影响审查标准进行评估，被认为值得支持。

The Internet of Things (IoT) refers to the ecosystem of computing devices (e.g. cellphones, household appliances, vehicles, etc.) connected to each other via the internet. The goal of the Pervasive Personalized Intelligence (PPI) Center is to enable the next generation of IoT systems to be more proactive and personalized without compromising security and privacy, thereby improving decision making, increasing efficiency and enabling new types of computing applications. This proposal will study the feasibility of establishing a new PPI site at the University of Michigan-Dearborn (UM-Dearborn) in collaboration with industry and the two PPI sites at the University of Colorado Boulder (CU) and Oregon State University (OSU).Researchers at UM-Dearborn will investigate foundational research under three thrusts: Intelligent Transportation, Smart Energy, and Software Engineering Intelligence. First, under the intelligent transportation thrust, investigators will develop new algorithms and techniques enabling personalized and context-aware maneuvering of connected and automated vehicles via artificial intelligence, i.e., intelligent personalization. Second, for the smart energy thrust, investigators will create new techniques that enable novel AI-assisted personalized demand response technologies for electric prosumers. Finally, under the software engineering intelligence thrust, investigators will explore methods to help software developers to build optimized PPI applications that can be efficiently and securely deployed and evolved in cyber-physical systems.Advances in IoT impact many facets of everyday life. The PPI Center explores research thrusts in Artificial Intelligence, Computing Systems, Programmability, Energy, Visualization, and Computer Security and Privacy that are broadly applicable beyond IoT systems to a variety of other scientific domains such as environmental science, agriculture, biology, transportation, and robotics. The research outputs will be disseminated through publications, research presentations to industry and academia, open-source code and industrial collaborations. The PPI Center is committed to broadening participation in computing through, for example, summer schools for industry engineers and K-12 outreach through partnerships with UM-Dearborn’s programs. To improve the retention of underrepresented populations, the proposed Center will work with technology associations to showcase PPI applications that help humanity.The PPI Center will maintain a publicly accessible project repository that will store open-source code, publicly available datasets, papers and other research outputs that can be disseminated to the general public on ppicenter.org.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.