نوع مقاله : مقاله پژوهشی
عنوان مقاله English
نویسندگان English
Controlling single-rotor tail-sitter unmanned aerial vehicles poses significant challenges due to highly nonlinear dynamics, severe aerodynamic couplings, and kinematic singularities at pitch angles near 90 degrees. This paper proposes an integrated, robust guidance and control architecture for hover and 3D trajectory tracking. To address these challenges, a cascaded hybrid control framework based on a 6-DOF model is developed. The outer loop utilizes a quaternion-based vectorial formulation to generate singularity-free reference attitudes, entirely resolving the gimbal lock issue inherent in Euler angles. The inner loop employs an Incremental Nonlinear Dynamic Inversion controller. By feeding back angular accelerations, this algorithm minimizes reliance on complex analytical aerodynamic models and guarantees robustness against uncertainties. Furthermore, a Weighted Least Squares control allocator dynamically distributes commands among aerodynamic surfaces and the propeller’s swashplate, utilizing real-time penalties. Command filters are embedded to attenuate sensor noise and strictly enforce actuator saturation limits. Extensive 6-DOF simulations under severe parametric uncertainties (30% modeling error) and continuous Dryden wind turbulence validate the system’s performance. Comparative results with the classical Nonlinear Dynamic Inversion demonstrate the proposed strategy’s superiority in disturbance rejection, elimination of high-frequency actuator chattering, and ensuring smooth, stable transitions.
کلیدواژهها English