Flight dynamics model identification of a meso-scale twin-cyclocopter in hover

Abstract

In this paper the flight dynamics of a 33-gram twin-cyclocopter is analyzed via deriving a Linear Time Invariant (LTI) dynamics model from flight test data. The twin-cyclocopter is a novel micro air vehicle that uses two co-rotating cycloidal rotors to generate thrust and a coaxial nose rotor to counteract the reaction torque and provide additional thrust. During flight tests, perturbation maneuvers were performed about the hovering state to excite different modes and a 3D motion capture system collected attitude and position data. The data was used to extract a bare airframe LTI model linearized about the hovering state using time-domain system identification techniques. The model demonstrated that the roll and yaw modes are gyroscopically coupled with stable high-frequency and low-frequency modes. Comparing the two different yaw control methods: thrust vectoring of the cycloidal rotors and differential torque of the coaxial nose rotor, the former was more effective.