<![CDATA[The orientation angle of a quadrotor UAV can be estimated from gyroscope and accelerometer data. Orientation can be predicted from gyroscope data under static or dynamic conditions, but the predicted value has accumulated errors. Meanwhile, orientation can also be calculated from accelerometer data, but only correct if the sensor is in a static state. To get a more precise orientation angle, the orientation predicted from the gyroscope data and the orientation calculated from the accelerometer data were fused using a Kalman filter. Determination of the condition of the sensor using a threshold value that is applied to the covariance of the acceleration data. in this study, the zero-acceleration compensation algorithm is used so that when the sensor is static, the orientation angle is calculated from the accelerometer. The use of this algorithm can increase the accuracy of the quadrotor orientation for roll angle to 96.84% and pitch angle to 98.91%.]]>
