Long Term Quadrotor Stabilization

The work of this thesis focuses on the IMU and getting the best performance possible out of the IMU to achieve better long term stability and a better navigation solution. This is done in two ways. First, the IMU accelerometer output is examined to determine if it is possible to use accelerometers to determine attitude. If the quadrotor is stationary or moving at constant velocity, the roll and pitch angles can be determined. Additionally, the accelerometers can be used to determine angular accelerations and angular rates which are integrated to determine heading. The second approach models the quadrotor and uses the models in Kalman Filters along with the IMU measurements to determine the best possible navigation solution.