Learning the Oscillations: A Novel Quaternion-Based Flapping Wing Flight Attitude Estimation Method Using MARG Sensors

Severe instantaneous oscillations in flapping wing aerial vehicle (FWAV) flight can exacerbate its attitude estimation problem. This paper presents a novel quaternion-based flapping-wing-flight attitude estimation algorithm. Unlike existing algorithms, the proposed one only uses low-cost magnetic angular rate and gravity (MARG) sensors, and can learn both the period lengths and the patterns of the oscillations online, even if the flapping wing frequency drifts rapidly. We conduct high-fidelity numerical simulations to validate the proposed algorithm. The obtained results demonstrate the obvious superiority of the proposed algorithm over existing ones.