SCC-DG: Digital Twin and AI-Infused Drones for Energy Retrofitting in Residential Envelopes

U.S. households spent on average 5.6% of their income on energy, with 40% of energy losses through the envelope. Homes with deteriorated insulation and aging structure face disproportionately high energy costs. Retrofitting and weatherization are key strategies for enhancing building performance and resilience against environmental stress. However, traditional energy audits and retrofitting methods are labor-intensive, costly, and often inaccessible to the populations who need them most. This project addresses these challenges by developing an advanced, autonomous drone system equipped with thermal imaging and Artificial Intelligence (AI) that can quickly and accurately detect energy loss in building envelopes. The drone system will work in coordination with Digital Twin (DT) – an interactive, informative 3D model that helps visualize and plan retrofitting strategies at multiple scales. By partnering with local community organizations, the project ensures that these technologies are not only technically effective but also socially acceptable, accessible, and impactful. Through interdisciplinary collaboration in building science, robotics, AI, DT, and community-based research, the project aims to transform weatherization practices and promote energy retrofitting through high-tech innovation. This project proposes integrating embodied AI and DT into drone systems to formulate an autonomous, intelligent robotic system that streamlines weatherization assessments and energy retrofitting planning in residential buildings for multi-stakeholders. The project will assess technical, regulatory, and community requirements to prepare for the development and implementation of this system, focusing on three key components: 1) Embodied AI-Driven Autonomous Sensing Drone: Study embodied AI, edge/cloud computing, and human-robot interaction methods to enable real-time, autonomous, and intelligent navigation of drones within complex building environments. 2) DT-Enabled Interactive Retrofitting Analysis: Explore bi-directionally interaction between drones and DT to support precise on-site surveying for retrofitting analysis, design, and planning. 3) Socio-Technical Feasibility and Scalability: Identify and address legal, ethical, and economic considerations to ensure that the deployment of the DT and AI-infused drone systems is feasible, privacy-conscious, and scalable across various community settings. The project involves co-design sessions, community focus groups, and pilot demonstrations to ensure that the technologies are aligned with the needs of stakeholders such as weatherization officials, community agencies, and homeowners. The research team will evaluate usability, scalability, and interpretability to guide future full-scale implementation. This interdisciplinary initiative spans civil engineering, robotics, computer science, and community engagement, aiming to produce transformative, deployable solutions that advance residential building’s sustainability and resilience. 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.