Geomagnetic data gathering by micro-rovers is gaining momentum both for future planetary exploration missions and for terrestrial applications in extreme environments. This paper presents research into the integration of a planetary micro-rover with a potassium total-field magnetometer. The 40 kg Kapvik micro-rover is an ideal platform due to an aluminium construction and a rocker-bogie mobility system, which provides good manoeuvrability and terrainability. A light-weight GSMP 35U (uninhabited aerial vehicle) magnetometer, comprised of a 0.65 kg sensor and 0.63 kg electronics module, was mounted to the chassis via a custom 1.21 m composite boom. The boom dimensions were optimized to be an effective compromise between noise mitigation and mechanical practicality. An analysis using the fourth difference method was performed estimating the magnetic noise envelope at ±0.03 nT at 10 Hz sampling frequency from the integrated systems during robotic operations. A robotic magnetic survey captured the total magnetic intensity along three parallel 40 m long lines and a perpendicular 15 m long tie line over the course of 3.75 h. The total magnetic intensity data were corrected for diurnal variations, levelled by linear interpolation of tie-line intersection points, corrected for a regional gradient, and then interpolated using Delaunay triangulation to lead a residual magnetic intensity map. This map exhibited an anomalous linear feature corresponding to a magnetic dipole 650 nT in amplitude. This feature coincides with a storm sewer buried approximately 2 m in the subsurface. This work provides benchmark methodologies and data to guide future integration of magnetometers on board planetary micro-rovers.

With the recent proliferation of unmanned aerial vehicles for geophysical surveying, a novel opportunity exists to develop unmanned ground vehicles in parallel. This contribution features a study to integrate the Husky A200 robotic development platform with a GSMP 35U magnetometer that has recently been developed for the unmanned aerial vehicle market. Methods to identify and reduce the impact of magnetically noisy components on the unmanned ground vehicle platforms are discussed. The noise generated by the platform in laboratory and gentle field conditions, estimated using the fourth difference method for a magnetometer–vehicle separation distance of 121 cm and rotation of the chassis wheels at full speed (1 m/s), is ±1.97 nT. The integrated unmanned ground vehicle was used to conduct two robotic magnetic surveys to map cultural targets and natural variations of the magnetic field. In realistic field conditions, at a full speed of 1 m/s, the unmanned ground vehicle measured total magnetic intensity over a range of 1730 nT at 0.1 m spatial resolution with a productivity of 2651 line metres per hour.

With the recent proliferation of unmanned aerial vehicles (UAV) for geophysical surveying a novel opportunity exists to develop unmanned ground vehicles (UGV) in parallel. This research presents a pilot study to integrate two UGVs, the Kapvik planetary micro-rover and a Husky A200 robotic development platform, with a GSMP 35U magnetometer that has recently been developed for the UAV market. Magnetic noise levels generated by the UGVs in laboratory and field conditions are estimated using the fourth difference method and, at a magnetometer-UGV separation distance of 121 cm, the Kapvik micro-rover was found to generate a noise envelope ± 0.04 nT whereas the noisier Husky UGV generated an envelope of ± 3.94 nT. The UGVs were assessed over a series of successful robotic mapping missions which demonstrated their capability for magnetic mapping, and their productivity and versatility in field conditions.iii Acknowledgements I would first like to thank my supervisor, Dr. Claire Samson, for her tireless support and confidence in the development of this project from start to finish. Dr.Samson's expertise in applied geophysical methods, skill as an editor, and focus on project goals ensured that this experience would be a success. For allowing me the freedom and opportunity to explore research blending the fields of geophysical exploration and robotics, I am eternally grateful. I would like to acknowledge the support of Dr. Alex Ellery, of Carleton University's Mechanical and Aerospace department, for generously providing access to the Kapvik planetary micro-rover and Husky A200 robotic development systems for the duration of this project. Dr. Ellery's insight and advice in discussion of designs and rover integration methodology proved very helpful. I would also like to offer a sincere thank you to GEM Systems Inc. for the loan of the GSMP 35U magnetometer which was a cornerstone of this project. The technical support and knowledgeable team at GEM were a delight to work with. A special thank you to Blair Walker and Mike Wilson for their availability for troubleshooting rovermagnetometer integration issues. To David Boteler, Benôit St-Louis, Lorne McKee, Nathan Olfert, and the Geomagnetism team at NRCAN, thank you for welcoming me at the Ottawa Geomagnetic Observatory over the course of my research. Your expertise in data iv analysis techniques, technical support, and insightful discussions of magnetic methodologies made a large contribution to the successful completion of this project.