SCC-PG: Towards cybernetic buildings: integrated intelligent sensing to create responsive environments

SCC-PG：迈向控制论建筑：集成智能传感以创建响应式环境

• 批准号: 1951952
• 负责人: Junsong Yuan
• 金额: $ 15万
• 依托单位: SUNY at Buffalo
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2020
• 资助国家: 美国
• 起止时间: 2020-10-01 至 2022-09-30
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1951952&HistoricalAwards=false
• 关键词: SCC; PG; Towards; cybernetic; buildings

The essential building blocks of a community are buildings. Like the larger community, buildings have to accommodate all the needs of a diverse group of people. Some "smart" systems are employed in the built environment, but they are designed to facilitate operations and improve efficiency. Do these systems benefit the inhabitants of the whole community, especially marginalized groups who often do not have a say in shaping policy and practices related to building management? In a democratic society, connected communities must be safe, comfortable, healthy, easy to use, and support productivity, contributing to the advancement of a just society for all. To meet these needs, a truly smart system should be cybernetic: it can incorporate feedback for self-regulation toward desirable goals. This research will develop and test an integrated platform to collect feedback on building performance, communicate that information to all building users, and create an online community through which members can influence the operation of building services and facilities. A proactive intelligent, or cybernetic system that engages the community of building users will provide the data to evaluate connected community strategies from a social perspective.The system will leverage smart phones and commercial augmented reality equipment to develop an inexpensive system of feedback and control that could be implemented in any building. It will include a sensor array for innovative building performance measurement, a set of dashboards for communicating and controlling the status of systems, and automated participatory processes that assess and communicate outcomes like satisfaction, comfort, and social engagement to facilities managers. The system will combine technologies from sensing, artificial intelligence, psychophysical research and social media. Implementing such systems on a large scale will provide data to create benchmarks for performance, identify unknown health threats, and understand the design features that contribute to productivity, satisfaction and community building. Once a working system has been implemented successfully for buildings, it can be scaled up for community wide use.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.

社区的基本构件是建筑物。与更大的社区一样，建筑必须满足不同人群的所有需求。一些“智能”系统被用在建筑环境中，但它们的设计是为了方便操作和提高效率。这些制度是否惠及整个社区的居民，特别是在制订大厦管理政策和做法方面往往没有发言权的边缘人士？在民主社会中，相互连接的社区必须安全、舒适、健康、易于使用，并支持生产力，为促进人人享有公正的社会作出贡献。为了满足这些需求，一个真正的智能系统应该是控制论的：它可以纳入自我调节的反馈，以实现理想的目标。这项研究将开发和测试一个综合平台，以收集关于建筑性能的反馈，将这些信息传达给所有建筑用户，并创建一个在线社区，成员可以通过该社区影响建筑服务和设施的运营。一个积极主动的智能或控制系统将从社会的角度提供数据，以评估互联社区的策略。该系统将利用智能手机和商业增强现实设备来开发一种廉价的反馈和控制系统，可以在任何建筑中实施。它将包括一个用于创新建筑性能测量的传感器阵列，一组用于沟通和控制系统状态的仪表板，以及用于评估和向设施经理传达满意度、舒适度和社交参与度等结果的自动化参与式流程。该系统将结合传感、人工智能、心理物理研究和社交媒体等技术。大规模实施这样的系统将提供数据，以创建性能基准，识别未知的健康威胁，并了解有助于提高生产率、满意度和社区建设的设计特点。一旦工作系统在建筑物中成功实施，它就可以扩大到社区广泛使用。这一奖项反映了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。

项目成果

Rethinking Soft Labels for Knowledge Distillation: A Bias-Variance Tradeoff Perspective

• 发表时间: 2021-02
• 期刊: ArXiv
• 作者: Helong Zhou;Liangchen Song;Jiajie Chen;Ye Zhou;Guoli Wang;Junsong Yuan;Qian Zhang

Personalized Prediction of Indoor Comfort Using Graph Convolutional Matrix Completion

• DOI: 10.1109/mipr54900.2022.00053
• 发表时间: 2022-08
• 期刊: 2022 IEEE 5th International Conference on Multimedia Information Processing and Retrieval (MIPR)
• 作者: Junyi Liu;E. Naidu;Jialian Wu;Shira Gabriel;E. Steinfeld;Junsong Yuan

Robust Knowledge Transfer via Hybrid Forward on the Teacher-Student Model

• DOI: 10.1609/aaai.v35i3.16358
• 发表时间: 2021-05
• 作者: Liangchen Song;Jialian Wu;Ming Yang;Qian Zhang;Yuan Li;Junsong Yuan

Track to Detect and Segment: An Online Multi-Object Tracker

• DOI: 10.1109/cvpr46437.2021.01217
• 发表时间: 2021-03
• 期刊: 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
• 作者: Jialian Wu;Jiale Cao;Liangchen Song;Yu Wang;Ming Yang;Junsong Yuan