Eventos Anais de eventos
COBEM 2021
26th International Congress of Mechanical Engineering
SYNTHETIC ESTIMATION OF ANGLE OF ATTACK AND SIDESLIP USING A MANEUVER RECONSTRUCTION METHOD FOR FLIGHT TESTING OF AN UAV
Submission Author:
Andrew Gomes Pereira Sarmento , SP
Co-Authors:
Andrew Gomes Pereira Sarmento, Alain Souza, Luiz Carlos Góes, ROBERTO GIL ANNES DA SILVA
Presenter: Andrew Gomes Pereira Sarmento
doi://10.26678/ABCM.COBEM2021.COB2021-2163
Abstract
With the growing development of remotely piloted aircraft, there is a need to identify aerodynamic parameters for future use to optimize control loops, making it possible to carry out autonomous missions. Due to the high cost of anemometric instruments and the low load capacity of the aircraft, there was a need to study methods for estimating the main anemometric data used in the identification process, the angles of attack, and sideslip. The main objective of this research is to estimate the angles of attack and sideslip of an Unmanned Aircraft Vehicle (UAV) and to observe the quality of this estimate comparing with the results measured by the instrumentation on board. The platform used was the UAV C2, equipped with the MicroPilot LRC 2 autopilot system for the provision of inertial data, GNSS, and relative speed, as well as a probe for the supply of data on the angles of attack and sideslip. The estimation of the angles comes from the maneuver reconstruction using the equations of the dynamic model. Once the attitude variables, angular velocities, and accelerations are effectively measured, rewrite the equations of the dynamic model as a function of the variables measured in the gravitational center of the aircraft. In-flight data are implemented in the equations for numerical integration, but due to bias errors in the accelerometers and drift in the gyroscopes, after the integral of the values from the inertial sensors, the results obtained show a drift error. The compensation for drift errors is made by comparing the main angular velocity variables related to the estimated data. When the stretch does not show variations in the angular velocities, drift on the estimated data is considered purely. The expected results are the estimates with low root mean squared (RMS) error about the measured data. At the end of this work, we want to conclude the effectiveness of the method for use in identifying aerodynamic parameters using minimizations in the time or the frequency domain of the UAV after flight tests.
Keywords
air data estimation, data reconstruction, Flight Test, UAV
