An Autonomous Drone Framework for Real-time 3D Construction Monitoring Using Photogrammetry and IoT Technologies

Abstract

This paper presents an innovative framework for autonomous 3D construction site monitoring using drones, photogrammetry, and Internet of Things (IoT) technologies. The proposed system captures 2D images of construction sites via a custom-built drone, which transmits the images in real-time to cloud services for immediate processing. Using Amazon Web Services (AWS), the images are automatically reconstructed into detailed 3D models, enabling remote monitoring and progress tracking. The system leverages Industry 4.0 technologies to facilitate cloud-based data analysis, offering a scalable and efficient solution for construction project management. We validated the framework through real-world testing at a construction site in Subang Jaya, Malaysia, where two flight missions were conducted, capturing 79 and 158 images, respectively. The 3D reconstruction was successfully performed with improved processing time on higher-specification AWS machines. Results showed that the system was able to reconstruct accurate 3D models, with Flight #2 achieving smoother surfaces compared to Flight #1, despite variations in image quality. The integration of Building Information Modeling (BIM) with the 3D models is proposed to automate the comparison of construction progress with design specifications. This approach has the potential to enhance the precision, accessibility, and efficiency of construction monitoring, particularly in environments with limited access.