The implementation of fuzzy logic control for attitude estimation of UAVs

This paper describes an attitude estimation of a UAV based on low-cost accelerometer and gyroscopes. This attitude is represented by its roll and pitch angles. The accelerometer can measure the angles between the actual gravity vector and the measured gravity related to the roll and pitch of UAV. The first estimation is based on accelerometer which we can take the advantage of fuzzy logic control implemented with Kalman filter to estimate the raw data of accelerometer. The second one is based on gyroscopes by integration of the differential equations relating to the UAV´s attitude and its instantaneous angular velocity. The results of the two estimations are compensated in order to take advantage on both of them. The experimental results show that this system can accurately estimate the attitude of the UAV including the testing on the vibration platform.