Precision positioning in unmanned aerial geophysics

This research investigates the implementation of precision GPS in Unmanned Aerial Vehicles (UAV) for use in geophysical exploration. The prompt for this research was the Desert Fireball Network’s meteorite recovery program, where an advanced impact site prediction system is followed by manual search and recovery. A small, automated, search vehicle is needed to explore the likely impact zones, which would require precise and accurate positioning in conjunction to its sensor capabilities. This thesis presents a Kalman filter implementation to improve and interpolate positioning during post-processing. This thesis also presents a sub-2kg UAV magnetometer system utilising an RTK GPS to achieve centimetric positioning. An RTK GNSS module was integrated with an Arduino microcontroller for acquisition of in-house magnetometer gradiometer data. Results are presented for two field trials, testing both positioning and magnetometer performance. The magnetometer performance was limited, particularly due to flight effects and sensitivity, however under ideal conditions the system would be capable of locating a meteorite sample. Dropout of DGPS during flight did occur during surveys, but the Kalman filter was successful in ameliorating the effects. This preliminary analysis is promising, and builds the foundation for future multispectral research for weed management in KNP, and provides a foundational methodology for spectral pre-assessment in general.