Hardware Implementation of Attitude Estimation Methods Using Multiple GPS Receivers

Abstract

This paper deals with the attitude estimation issue for navigation of Unmanned Aerial Vehicles (UAVs). We propose a new multi-GPS sensor fusion, based on Extended Kalman Filter (EKF) and Smooth Variable Structure Filter (SVSF), using an Inertial Measurement Unit (IMU) and three GPS receivers. Two methods; the Direct Computing Method (DCM) and the Least Square Method (LSM) are used to estimate the full attitude of the UAV. Performance and robustness are tested with numerical simulation. Experimental validation is carried out on fixed-wing aircraft to evaluate the capabilities of the proposed approaches in attitude estimation. Through this framework, it is proven that subdegrees order of estimation error can be guaranteed when real dynamic situations are applied to the aircraft.