SCC-CIVIC-PG Track A: Full Building Scans for Targeted Micro-retrofits using Drones, Radars, and Deep Learning

SCC-CIVIC-PG 轨道 A：使用无人机、雷达和深度学习进行全面建筑扫描以进行有针对性的微型改造

• 批准号: 2228568
• 负责人: Chen Feng
• 金额: $ 5万
• 依托单位: New York University
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2022
• 资助国家: 美国
• 起止时间: 2022-10-01 至 2023-08-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2228568&HistoricalAwards=false
• 关键词: SCC; CIVIC; PG; Track; Full

Poor air and moisture sealing in building envelopes contribute to worse buildings emissions and occupant health outcomes. Finding moisture using thermal scans can be a laborious and environmentally constrained process. Our contribution in ground penetrating radar involves using machine learning to find anomalous areas of interest in radar scans. We will be able to find trapped and hidden moisture that is not detectable with current non destructive testing techniques. By partnering with the City of New York and NGOs like District 2030, this project will (a) research the applicability of radar scans on different types of building facades, (b) create real, actionable outcomes that will improve the quality of life for residents of disadvantaged communities in NYC by creating better buildings through improving building envelopes with targeted micro-retrofits, (c) create a product that is beneficial to the new construction, the existing construction, the building insurance, the property management, and the building engineering industries. This project will enhance US competitiveness, produce new products, bolster economic growth, and benefit society at large. The findings from this project have the potential to scale and produce similar positive outcomes across the nation.Poorly maintained building envelopes exacerbate building greenhouse gas emissions and cause quality of life problems. We propose a non-invasive integrated solution to locate and document moisture intrusion, thermal bridges, and air leaks to diagnose building envelope issues. The system identifies and quantifies common envelope defects and applies long-wave radar and deep learning to detect hidden deep moisture penetration and other major envelope defects. This project aims to perform inspections on public and private buildings in New York City, specifically in low-income communities. We will work with community organizations and local government departments already carrying out this work and enhance their efforts through AI and robotics technologies being actively developed in the PI/Co-PI’s labs. This project will improve the quality of life for residents in low-income communities by creating more comfortable and usable buildings. With this system, it is possible to perform low-cost, targeted micro-retrofits to address envelope issues. We will either address building envelope issues directly through low-cost targeted micro-retrofits or we will propose a list of retrofits to our partners when that is not possible. This project is in response to the Civic Innovation Challenge program—Track A. Living in a changing climate: pre-disaster action around adaptation, resilience, and mitigation—and is a collaboration between NSF, the Department of Homeland Security, and the Department of Energy.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.

建筑围护结构中的空气和湿气密封不良会导致建筑物排放和居住者健康状况恶化。使用热扫描查找湿气可能是一个费力且受环境限制的过程，我们在探地雷达方面的贡献涉及使用机器学习来查找雷达感兴趣的异常区域。通过与纽约市和 District 2030 等非政府组织合作，我们将能够发现当前无损检测技术无法检测到的残留和隐藏的水分，该项目将 (a) 研究其适用性。对不同类型的建筑立面进行雷达扫描，(b) 创造真实、可行的结果，通过有针对性的微改造改善建筑围护结构，创造更好的建筑，从而提高纽约市弱势社区居民的生活质量，(c) 创造一项有利于新建建筑、现有建筑、建筑保险、物业管理和建筑工程行业的产品。该项目将增强美国的竞争力，促进新产品的发展，促进经济增长，造福整个社会。从这个项目有潜力规模化并在全国范围内产生类似的积极成果。维护不善的建筑物会加剧建筑物温室气体排放并导致生活质量问题。我们提出了一种非侵入性集成解决方案来定位和记录湿气侵入、热桥和空气泄漏，以诊断建筑物。该系统识别和量化常见的围护结构缺陷，并应用长波雷达和深度学习来检测隐藏的深层湿气渗透和其他主要围护结构缺陷。该项目旨在对纽约市的公共和私人建筑进行检查。我们将为低收入社区工作。社区组织和地方政府部门已经开展了这项工作，并通过 PI/Co-PI 实验室积极开发的人工智能和机器人技术加大力度。该项目将通过创建低收入社区居民的生活质量。借助该系统，可以进行低成本、有针对性的微型改造来解决围护结构问题，我们将直接通过低成本提出有针对性的微型改造来解决建筑围护结构问题，或者我们将列出清单。对我们的改造该项目是为了响应公民创新挑战计划——A.生活在不断变化的气候中：围绕适应、复原力和缓解的灾前行动——并且是 NSF 与美国农业部之间的合作项目。国土安全部和能源部。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。

项目成果

DeepGPR: Learning to Identify Moisture Defects in Building Envelope Assemblies from Ground Penetrating Radar

DeepGPR：学习通过探地雷达识别建筑围护结构组件中的潮湿缺陷

• DOI: 10.22260/isarc2023/0075
• 发表时间: 2023
• 期刊: International Symposium on Automation and Robotics in Construction (ISARC
• 作者: Sher, Bilal Ali;Feng, Chen
• 通讯作者: Feng, Chen

Marker-based Extrinsic Calibration for Thermal-RGB Camera Pair with Different Calibration Board Materials

具有不同校准板材料的热 RGB 相机对基于标记的外部校准

• DOI: 10.22260/isarc2023/0066
• 发表时间: 2023
• 期刊: International Symposium on Automation and Robotics in Construction (ISARC
• 作者: Sher, Bilal Ali;Xu, Xuchu;Chen, Guanbo;Feng, Chen
• 通讯作者: Feng, Chen