A Robust Framework For Dynamic Soaring Trajectory Generation Accounting For Variation Of Aerodynamic Derivatives With Flight Conditions

Dynamic soaring is a biologically inspired flight technique that harnesses wind shear for energy-efficient, sustained flight, as demonstrated by seabirds such as albatrosses. This capability holds significant promise for unmanned aerial vehicles (UAVs) tasked with long-endurance missions in challenging environments. Existing approaches to trajectory generation have largely relied on direct optimal control formulations or differential-flatness-based methods, often under the simplifying assumption of constant aerodynamic derivatives. Such assumptions, however, limit accuracy and robustness when UAVs encounter variations in flight conditions. This paper presents a robust six-degree-of-freedom (6-DOF) trajectory generation framework that enhances the flatness-based formulation by incorporating sideslip and the complete set of stability and control derivatives. The study begins with a comparative analysis between the original and improved formulations using Floquet stability theory and control input profiles. To examine the implications of the constant-derivative assumption, trajectories generated under fixed aerodynamic derivatives are compared against steady aerodynamic forces and moments obtained from a Vortex Lattice Method (VLM) solver. A 6-DOF UAV model stabilized by a Linear Quadratic Regulator (LQR) is then simulated in dynamic soaring orbits, revealing divergence and highlighting the limitations of the constant-derivative approach. To overcome these limitations, the framework integrates lookup tables generated via VLM that capture the variation of aerodynamic derivatives with airspeed, angle of attack, sideslip, and altitude. The resulting formulation produces dynamically consistent and stable trajectories that better reflect nonlinear aerodynamic effects and turbulence. Simulation results demonstrate that accounting for derivative variation significantly improves trajectory stability and robustness, offering a more realistic and reliable methodology for dynamic soaring applications. Keywords: Dynamic Soaring, Differential Flatness, Aerodynamic Derivatives, Vortex Lattice Method, Floquet Stability