Investigating the Potentials of Small Uncrewed Aircraft Systems (sUAS) for Gravel Road Dust Monitoring

Document Type

Conference Proceeding

Publication Date

1-1-2026

Abstract

Gravel/unsurfaced/unpaved roads make up approximately 35% of all roadways in the United States, and in states such as Iowa, nearly 60% remain unpaved. These networks require consistent monitoring to maintain optimal performance and extend their service life. Additionally, dust clouds produced by moving vehicles typically reduce visibility, raise environmental concerns, and pose respiratory risks. The United States Army Corps of Engineers recommends assessing dust severity using the dust cloud generated by a vehicle traveling at 25 miles per hour. Still, this approach places assessors in unsafe conditions and is not practical for long routes or high dust levels. Earlier studies using vehicle-mounted cameras have faced issues with dust buildup and unstable mounts. To address these limitations, this study explores the use of sUAS for rating dust clouds at different speeds. Using data collected from three Iowa gravel roads, we adopted deep learning models to classify dust severity in sUAS images, achieving a weighted F1 score of 0.9467.

Publication Title

International Conference on Transportation and Development 2026 Transportation Safety and Emerging Technologies Selected Papers from the International Conference on Transportation and Development 2026

ISBN

[9780784487013]

Link to Full Text

Share

COinS