From Global GPS to Local Waypoints: A UTM-Based GPS-IMU Localization Method for VTOL UAV Navigation

Abstract

This study presents a deterministic UTM-based GPS–IMU localization framework for short-range waypoint navigation of a Vertical Take-Off and Landing (VTOL) unmanned aerial vehicle. Unlike conventional GNSS–IMU approaches that rely on tightly coupled probabilistic fusion algorithms, the proposed method emphasizes coordinate transformation efficiency for relative waypoint consistency under open-sky conditions. GPS latitude–longitude data from a standalone u-blox M10 receiver are converted into Universal Transverse Mercator (UTM) coordinates and transformed into a local Cartesian reference frame to enable efficient distance and bearing computation. Heading information is obtained from IMU yaw measurements embedded in the Pixhawk 6C flight controller. Experimental validation was conducted in an open-sky environment using five predefined waypoint positions per trial. Performance was evaluated using Root Mean Square Error (RMSE) and sample standard deviation metrics. The system achieved distance RMSE values of 0.030 m, 0.047 m, and 0.017 m across three scenarios, while bearing RMSE remained within 4–5°, satisfying the predefined benchmark (distance RMSE < 0.1 m; bearing RMSE < 5°). These results reflect short-range relative geometric consistency rather than absolute GNSS positioning accuracy. The findings demonstrate that deterministic UTM-based coordinate transformation combined with GPS–IMU heading estimation provides stable short-range waypoint consistency without requiring additional probabilistic fusion algorithms.