Model-Aided State Estimation of HALE UAV With Synthetic AOA/SSA for Analytical Redundancy

Abstract

This paper proposes a novel dynamic model-aided navigation filter to estimate the safety-critical states of a high-altitude long-endurance (HALE) UAV without measurement of angle of attack (AOA) and sideslip angle (SSA). The major contribution of the proposed algorithm is that the synthetic AOA and SSA measurements are newly formulated for analytical redundancy. In the proposed filter, aerodynamic coefficients and control signals are utilized along with inertial measurement unit (IMU), Global Positioning System (GPS), and pitot tube measurements to estimate the navigation states as well as the steady and turbulent effects of 3D wind using random walk (RW) and Dryden wind models, respectively. Flight test results of a HALE UAV demonstrated that the proposed algorithm yields accurate estimated airspeed, AOA, SSA, attitude, angular rates, and 3D wind states, demonstrating its effectiveness for analytical redundancy.