Kalman-based attitude estimation for an UAV via an antenna array

Accurate attitude estimation is crucial for Unmanned Aerial Vehicles (UAVs) in order to help them to perform several automated activities such as following a trajectory or landing. Most attitude and navigation algorithms rely on a inertial measurement unit (IMU), i.e., accelerometer and gyro sensors, and some auxiliary sensors, such as magnetometer. Recently antenna array based communication systems have being installed in UAVs. This structure can be also applied for attitude estimation by computing the direction of arrival (DOA) of the line of sight (LOS) path between the base station and the UAV. In this paper, we present three main contributions. First, assuming that the yaw angle is already estimated or measured, we propose an analytical expression for computing the pitch and the roll angles by using only the DOA estimates, which shows a good performance only for high SNR scenarios. As a second contribution, we propose to apply the known TRIAD algorithm using as input the estimated DOAs of the LOS and magnetometer measurements. This provides significantly better results than the first contribution in low SNR scenarios. As a third contribution, we propose to apply the Kalman filter on the gyro plus magnetometer plus estimated DOAs of the LOS. Due to the tracking capability of the Kalman, the attitude estimates become more robust against the noise influence.