Attitude and Heading Reference System for Small Unmanned Aircraft Collision Avoidance Maneuvers

This thesis describes the development of an Attitude and Heading Reference System (AHRS) to sense three-dimensional orientation for collision avoidance control in small unmanned aircraft. Unmanned aircraft are currently restricted to flight in designated airspace due to safety concerns of collision with manned aircraft. Therefore, collision avoidance is necessary to ensure the safety of both aircraft. Technical challenges, mainly in sensor limitations, restrict AHRS performance in attitude estimation during high-g maneuvers. Using sensor filtering techniques and a robust attitude representation, an AHRS suitable for collision avoidance is developed. Acceleration disturbances are reduced using estimates of non-gravitational accelerations including centripetal acceleration and model-based acceleration to improve gravity vector measurement during aircraft maneuvers. Simulation results with a variety of maneuvers deemed challenging for most AHRS are given showing accurate attitude estimates. Flight data from an existing commercial autopilot is compared with the results of the AHRS to demonstrate the validity of the solution with real flight data.