Research on simulation method of measuring pavement smoothness with UAV-borne LiDAR

Traditional pavement smoothness detection methods suffer from low efficiency and limited coverage, struggling to meet large-scale, high-precision inspection demands. This study proposes a UAV((Unmanned Aerial Vehicle))-LiDAR (Light Detection and Ranging)-based pavement quality evaluation method using simulated point cloud analysis to overcome the reliance on costly field data and environmental noise interference. Current approaches are constrained by high operational costs and challenges in suppressing complex noise (e.g. vegetation occlusion, equipment vibration). Our method constructs parametric models of four typical defects (cracks, bumps, pockmarked surfaces, honeycombs) in MATLAB, generates high-fidelity point clouds by integrating UAV dynamics and LiDAR sensor characteristics, and employs CloudCompare for noise suppression and defect analysis through TIN(Triangulated Irregular Network) and DSM(Digital Surface Model) models. Experimental results demonstrate the method's feasibility, achieving an 18.7% improvement in defect detection accuracy and a 42.3% faster processing speed compared to traditional techniques. This work provides a low-cost, high-precision solution for construction quality assessment in scenarios lacking measured data, supporting smart infrastructure maintenance. The proposed framework lays a foundation for future integration of dynamic attitude compensation and deep learning to enhance engineering applicability.