Robust reduced-attitude control of fixed-wing UAVs using a generalized multivariable super-twisting algorithm

Operation of fixed-wing unmanned aerial vehicles (UAVs) outside their nominal operating conditions require autopilots that can effectively compensate for highly uncertain aerodynamics, and coupled disturbances due to turbulent winds. In this paper, we propose to use a generalized multivariable super-twisting algorithm to solve the robust attitude control problem for fixed-wing UAVs. A sliding surface is designed, based on geometric methods, to perform reduced-attitude tracking while simultaneously stabilizing a turn rate based on the coordinated-turn equation. The reduced-attitude representation evolves on the unit two-sphere and is independent of the yaw/heading angle. The resulting control design is lightweight, has no singularities, and can be used with standard hierarchical control architectures for fixed-wing UAVs. The efficacy of the proposed design is demonstrated in a simulation study with highly turbulent conditions.