Evaluating the Accuracy of UAV-Obtained Horizontal and Vertical Coordinates in District Six, Cape Town.

EasyChair Preprint 15157

Abstract

The use of Unmanned Aerial Vehicles (UAVs) for geospatial data collection has grown significantly due to their ability to capture high-resolution imagery and generate accurate topographic models. This study focuses on evaluating the accuracy of UAV-obtained horizontal and vertical coordinates in District Six, Cape Town, a historically complex urban environment. The research investigates the precision of UAV-based geospatial data by comparing it with ground control points (GCPs) surveyed using traditional methods such as GNSS (Global Navigation Satellite System). Key metrics such as horizontal positional accuracy and vertical elevation accuracy were analyzed, considering factors such as flight parameters, sensor quality, and post-processing techniques. The results reveal that UAVs can produce highly accurate geospatial coordinates, with horizontal and vertical errors falling within acceptable thresholds for urban mapping projects. However, factors like terrain variation, GCP distribution, and image processing limitations can affect the overall accuracy. The study concludes that UAV technology is a reliable tool for geospatial mapping in urban settings like District Six, but certain challenges and limitations must be addressed to enhance accuracy in more complex environments.

This research contributes to the growing body of knowledge on UAV applications in urban geospatial studies and highlights the potential for future integration of UAV data into large-scale urban planning and development efforts.

Keyphrases: Horizontal Accuracy, Mean Absolute Error (MAE), Root Mean Square Error (RMSE), aerial surveying, geospatial data, vertical accuracy

@booklet{EasyChair:15157,
  author    = {Anthony Collins},
  title     = {Evaluating the Accuracy of UAV-Obtained Horizontal and Vertical Coordinates in District Six, Cape Town.},
  howpublished = {EasyChair Preprint 15157},
  year      = {EasyChair, 2024}}