Digital twins for green roofs and facade inspection using drones and multi-modal foundation models

AI for Sustainable Construction, University of Rostock, Justus-von-Liebig-Weg 2, 18059 Rostock, Germany

^* Corresponding author: nils.krause@uni-rostock.de

Abstract

Green roofs and facades are a nature-based way to improve the climate of buildings. However, they also add additional maintenance effort for keeping the plants healthy. We propose a novel automated digital twin approach to monitor the health of the plants using a combination of drones with Foundation Models to monitor and analyze the health of the plants. The process consists of three main components, a drone system for recording real-time data in an automated manner for different heights, a digital twin for storing and controlling the data, and a Multi-modal Foundation model (MMFM) for diagnosis of health issues and generating recommendations for solving the issues. This allows us to automatically identify and mitigate health issues of the plants and improve the maintenance of the miniature ecosystems of green roofs and facades to address climate change risk and the shortage of skilled workers.