Date of Graduation

8-2024

Document Type

Thesis

Degree Name

Master of Science in Civil Engineering (MSCE)

Degree Level

Graduate

Department

Civil Engineering

Advisor/Mentor

Wood, Clinton M.

Committee Member

Barry, Michelle L.

Second Committee Member

Hernandez, Sarah V.

Keywords

DEM; Drone; Geospatial; LIDAR; Survey; UAV

Abstract

Unmanned aerial systems (UAS) LiDAR was used to collect survey data for small-area projects, particularly bridge replacement projects. The project aimed to compare UAS LiDAR data with conventional surveying methods. For the project, five bridge sites were selected for UAS LiDAR and conventional survey data collection, which include Lincoln Bridge 1 (East), Lincoln Bridge 2 (West), Humnoke, Frenchman’s Bayou, and Mountain Home. Data for each site was collected during the 2021–2022 winter. A DJI M600 Pro and Phoenix LiDAR Systems AL3-16 LiDAR unit were used for the UAS LiDAR measurements. The conventional survey equipment used included a global navigation satellite system (GNSS) real time kinematics (RTK) system and a robotic total station. LiDAR missions were planned using a combination of Google Earth, Phoenix LiDAR Systems’ Flightplanner, and Litchi Mission Hub. Flights were conducted at an altitude of 45 meters above ground level (AGL), with 50 percent side lap, ensuring comprehensive coverage. Weather and environmental conditions were carefully considered for optimal data collection. The LiDAR data were processed using Phoenix LiDAR System’s LiDARMill online software service. The processing involved GNSS data correction, trajectory optimization, and LiDAR data alignment and classification. Additionally, LAS tools were employed to enhance ground classification, reduce noise, and generate accurate digital elevation models. The accuracy of UAS LiDAR was assessed by comparing the UARK and ARDOT LiDAR datasets and ground survey data collected using GNSS and total station checkpoints for hard and soft surfaces. Errors between the UAS LiDAR and checkpoints were compared using aerial photos, bar graphs, and box and whisker plots, and direct raster comparisons were made whenever multiple LiDAR datasets were available. Overall, errors of up to approximately 1.0 inch can be expected for hard surfaces, with potential for both over- and under-predictions. An overprediction error of 1.0 inch to 3.5 inches can be expected for grass surfaces, depending on grass height. Larger overprediction errors of 2.0 inches to 5.0 inches can be expected for tall grass areas, while errors ranging between 2.0 inches and 7.0 inches can be expected for tree areas. The cost savings analysis showed that UAS LiDAR demonstrated an average cost reduction of $1,195.41 per project compared to helicopter LiDAR, resulting in a 20 percent reduction in the cost of collecting data for small-area bridge projects. In addition, UAS LiDAR exhibited an average cost reduction of $10,539.18 per bridge project compared to conventional survey methods, resulting in a 25 percent reduction in the cost of surveys for small-area bridge projects. Overall, UAS LiDAR proves to be a valuable tool for obtaining elevation data, with good accuracy for hard surfaces and expected variations for soft surfaces. Despite the limitations, the cost savings analysis puts UAS LiDAR surveying a clear advantage over both helicopter LiDAR and conventional surveying methods for small-area bridge projects.

Download

Share

COinS