Sun Sensor Geolocalization System: Field Testing and Navigational Application

This thesis presents a sun sensor and inclinometer-based geolocalization system intended for mobile robots. Using only the on-board sensors, combined with solar ephmeris information, terrestrial location and heading estimates are generated. This concept is validated using a post-processing approach, with field test data collected from an instrumentation package developed for this research. Simulations of this system are created to show potential performance with improved hardware. This sun sensor geolocalization system is then applied to an aided wheel odometry-based relative positioning framework. A novel algorithm, Fused Wheel Odometry, is proposed. This algorithm is designed to provide heading information during a mobile robot traversal, followed by an absolute localization update. Simulated data is used to show the improvements of this algorithm over typical aided wheel odometry methodologies. Potential applications of this geolocalization and aided wheel odometry system are explored, showing how this navigational approach could be incorporated into a typical rover mission.