The purpose of this paper is to design an output feedback control algorithm based on a reduced-order observer and a nonlinear disturbance observer (DO) for the estimation and rejection of unknown bounded disturbances with unknown frequency and magnitude.
For the trajectory tracking of the quadrotor, an integral sliding mode controller is initially designed. Subsequently, a nonlinear DO is developed to attenuate unknown bounded disturbances. Following this, a reduced-order observer is designed for state estimation. The control algorithm is then finalized by combining the controller, DO and reduced-order observer.
The findings of this study comprise a control algorithm devised through the integration of integral sliding mode control, a nonlinear DO and a reduced-order state observer. Unlike traditional state observers, where the output is used to estimate both output and output rates, this technique exclusively uses output position and attitude to estimate the output rates of position and attitude. This approach reduces the computation power demand on the controller. Additionally, the nonlinear DO effectively attenuates disturbances with unknown characteristics, such as frequency and magnitude, without necessitating prior knowledge of these parameters.
This paper introduces a criterion that uses measurable state variables, namely, position and attitude, to estimate the rate of change of these variables, specifically the rate of change of position and attitude. Additionally, another significant contribution of this paper is the design of a nonlinear DO to estimate exogenous disturbances, including repeated ramp disturbances and chirp disturbances characterized by variable frequency and magnitude.
