Attitude Optimal Control Based on Anti-Trembling Nonlinear Extended State Observer

This paper proposed an attitude control scheme accommodating strong nonlinearity and external disturbances. Small perturbation linearization method is used to simplify the mathematical model of plant. Based on the linearized model, an optimal attitude controller was designed. In order to compensate for the errors caused by linearization and external disturbances, an anti-trembling nonlinear extended state observer (ESO) has been composed in the control scheme. The control allocation method based on auto-encoder network achieve high-quality effect in the event of changes in dynamic parameters caused by attitude changes. Numerical examples show that the proposed method has good robustness against external disturbances.