Accurate Attitude Control for an Asymmetrical Hexacopter Under Exogenous Disturbance

Abstract

The distributed electric propulsion (DEP) vertical take-off and landing (VTOL) drone has the characteristics of both rotary-wing and fixed-wing aircraft. The dynamic model of its take-off and landing stage is similar to that of the multirotors. However, its large span makes it more susceptible to exogenous disturbances. Therefore, this paper leverages a modified uncertainty and disturbance estimator (MUDE) to dynamically estimate and compensate for the disturbance so that this type of drone has a relatively stable attitude motion in its VTOL stage. Considering the actuators of the drone in real flight are non-ideal, the motor dynamics is introduced into the controller design, its model is described as a first-order pulse with dead time (FOPDT) and the parameter was identified using a thrust test platform. In addition, the cascade general inverse (CGI) method is utilized to deal with the problem of actuator saturation in the control allocation for its asymmetrical layout. In order to verify the designed controller and control allocation algorithm, a small-scale asymmetrical hexacopter prototype was built, then the rotary platform experiments were carried out to fully exploit the attitude tracking performance of the proposed controller. A real flight test was also carried out to test the feasibility of the method. The experimental results show that the MUDE-based controller combined with the CGI control allocation scheme for our hexarotor prototype achieves accurate attitude tracking results under random exogenous disturbance.