Phase-parameterised gaussian process for predicting UAV aerodynamic loads in operational turbine wakes

This project develops and tests a control‑focused model that predicts how aerodynamic loads change for UAVs flying in the unsteady near‑wake of wind turbines. Because high‑fidelity fluid simulations are slow, the approach combines a free‑vortex‑wake representation with a phase‑parameterized Gaussian Process Regression (GPR) surrogate. Working in a coordinate system aligned with the wake, the surrogate takes advantage of the regular spatial and temporal patterns in the turbine‑generated velocity field. The GPR surrogate is then linked to a nonlinear, physics‑based rotorcraft flight‑dynamics model to estimate the unsteady aerodynamic forces on a quadrotor. This method can predict loads faster than real time, making it useful for flight control and trajectory planning. Support

for the project includes setting up the mathematics of the phase‑parameterized GPR, advice on coupling wake models with multi‑rotor dynamics, and creating validation procedures using wind‑tunnel data. The overall framework points to research that can improve UAV stability and safety during offshore wind‑turbine inspections.

Message: Hi FE, I need help with “Phase-parameterised gaussian process for predicting UAV aerodynamic loads in operational turbine wakes” in “Aerospace Engineering”