RTK and PPK GNSS Positioning Accuracy Assessment for UAVs Using a PPS-Synchronized Robotic Total Station

Accurate drone positioning is essential for applications such as photogrammetric surveying and environmental monitoring. Among differential GNSS techniques, Real-Time Kinematic (RTK) and Post-Processed Kinematic (PPK) are widely adopted to achieve centimeter-level positioning accuracy. However, their performance is typically assessed indirectly through the quality of derived products rather than by comparison with independent ground-truth measurements under dynamic flight conditions. This study presents a direct comparison of RTK and PPK positioning for drone applications using a robotic Total Station (TS) as an independent reference. The experimental platform is based on the Versatile External Synchronization and Telemetry Architecture (VESTA), which integrates differential GNSS receivers, an inertial measurement unit, and Coordinated Universal Time (UTC)-synchronized data logging. A synchronization procedure based on the GNSS Pulse-Per-Second (PPS) signal and the controlled motion of a TS prism was developed to accurately align GNSS and TS measurements. Flight experiments show that RTK and PPK achieve comparable accuracy when the reported positioning quality is optimal. However, for the tested short-baseline datasets and the RTKLIB-EX configuration, the PPK solution remained in float or degraded states for longer following GNSS degradation, resulting in larger transient errors than the real-time u-blox solution. These findings highlight that the choice between RTK and PPK should consider not only nominal positioning accuracy but also reconvergence behavior under dynamic operating conditions.