Publications / 2025 Proceedings of the 42nd ISARC, Montreal, Canada
Unmanned Aerial Vehicles (UAVs) have become indispensable for building inspections, offering safer, more efficient, and automated alternatives to traditional manual methods. However, current approaches often rely on predetermined inspection parameters, such as fixed distances, without adequately aligning with specific inspection goals or ensuring consistent data quality. This paper presents a novel methodology that addresses these challenges by integrating Building Information Modeling (BIM) and Ground Sampling Distance (GSD) into the inspection planning process. BIM is leveraged to automate the generation of Field of Views (FoVs) and viewpoints, significantly reducing human intervention and ensuring precise coverage of inspection areas. GSD is utilized to determine the optimal distance between the UAV and the surface, guaranteeing that collected data meets predefined objectives, such as detecting fine defects. To further enhance efficiency, a Genetic Algorithm (GA) is employed to optimize the flight path, minimizing travel distance while maintaining comprehensive coverage of all required areas. While the methodology is developed primarily for building inspections, it is scalable and adaptable to other infrastructure types, demonstrating its versatility in diverse scenarios. By combining BIM-based automation, GSD-driven planning, and advanced optimization techniques, this work provides a robust solution for automated, efficient, and high-quality UAV-enabled inspections for future innovations in automated inspection systems.