Dynamic Accuracy Improvement of a MEMS AHRS for Small UAVs

Abstract

Nowadays, unmanned aerial vehicles are widely used for various applications, with their sizes progressively decreasing as avionics are miniaturized. The development of microelectromechanical system (MEMS) technology has enabled miniaturization of avionics via application of the MEMS inertial measurement unit (IMU) to navigation sensors such as the attitude heading reference system (AHRS). However, perturbation or acceleration is known to cause low accuracy in the MEMS AHRS, which incorporates a critical sensor. Thus, in this paper, a method is proposed to improve the dynamic accuracy of a MEMS AHRS when an aircraft accelerates. The attitude calculations implement a quaternion method. The attitude correction of the AHRS algorithm entails the use of a Kalman filter, which is modified to improve the dynamic accuracy of the AHRS by adjusting the measurement noise covariance of the filter to change according to the maneuvering condition. The performance of the proposed algorithm was evaluated via simulation. For the simulation, actual flight data were acquired via a storage device that synchronizes the output of the reference sensor — a GPS-aided AHRS — and the IMU. Simulation of the proposed algorithm demonstrates that the proposed attitude estimation method yields results that are similar to the output of the reference sensor.