Year: 2026
Type: article
Abstract: Real-time monitoring of transportation infrastructure is crucial for ensuring the safety and operational efficiency of road networks. Traditional manual inspections involve high risks and low efficiency, whereas unmanned aerial vehicles (UAVs), with their high maneuverability and broad coverage, have gradually become an important data collection platform in the field of pavement defect detection. However, the large-scale application of UAVs in complex environments still faces technical bottlenecks such as environmental interference, inconsistent data quality, and slow processing of massive images. This paper systematically reviews the frontier progress of UAV-based pavement defect monitoring technologies in recent years: First, it outlines the limitations and challenges during the UAV field data collection phase; Second, it deeply analyzes the algorithmic evolution in office data processing, from image preprocessing and 3D pavement reconstruction to deep learning and semantic segmentation; Building upon this, the paper constructs a multi-dimensional evaluation system ranging from apparent 2D defects to hidden 3D defects and summarizes multi-modal fusion recognition methods, including visible light, LiDAR, infrared thermal imaging, and ground-penetrating radar (GPR). Finally, the paper discusses the limitations of current technologies regarding hardware battery life, legal privacy, and complex background interference, and points out the development potential of edge computing, UAV swarm collaboration, and cyber-physical systems in future intelligent road maintenance.
Author: Wang, Jianping
Institution: Hebei University of Technology
Language: English
Subfield: Civil and Structural Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: diamond
OpenAlex ID: W7215370917
