Machine learning enhanced real-time aerodynamic forces prediction based on sparse pressure sensor inputs

TL;DR

This paper tackles real-time, GNSS-free prediction of UAV aerodynamic forces using sparse surface pressure sensors. It combines a DEIM-based linear reduced-pressure reconstruction with a small neural-network correction to capture nonlinear dynamics and bridge gaps between simulations and experiments. Sensor placement is optimized via DEIM, and aerodynamic coefficients are computed by integrating the reconstructed pressure field, with the NN correcting residual errors to yield $C_l$ and $C_d$ that closely match ground truth. Demonstrated on URANS data for a 2D NACA0015 airfoil and a 3D drone, the DEIM+NN approach delivers fast, accurate lift and drag predictions with robustness to noise, offering a practical, GNSS-free tool for real-time UAV navigation and control.

Abstract

Accurate prediction of aerodynamic forces in real-time is crucial for autonomous navigation of unmanned aerial vehicles (UAVs). This paper presents a data-driven aerodynamic force prediction model based on a small number of pressure sensors located on the surface of UAV. The model is built on a linear term that can make a reasonably accurate prediction and a nonlinear correction for accuracy improvement. The linear term is based on a reduced basis reconstruction of the surface pressure distribution, where the basis is extracted from numerical simulation data and the basis coefficients are determined by solving linear pressure reconstruction equations at a set of sensor locations. Sensor placement is optimized using the discrete empirical interpolation method (DEIM). Aerodynamic forces are computed by integrating the reconstructed surface pressure distribution. The nonlinear term is an artificial neural network (NN) that is trained to bridge the gap between the ground truth and the DEIM prediction, especially in the scenario where the DEIM model is constructed from simulation data with limited fidelity. A large network is not necessary for accurate correction as the linear model already captures the main dynamics of the surface pressure field, thus yielding an efficient DEIM+NN aerodynamic force prediction model. The model is tested on numerical and experimental dynamic stall data of a 2D NACA0015 airfoil, and numerical simulation data of dynamic stall of a 3D drone. Numerical results demonstrate that the machine learning enhanced model can make fast and accurate predictions of aerodynamic forces using only a few pressure sensors, even for the NACA0015 case in which the simulations do not agree well with the wind tunnel experiments. Furthermore, the model is robust to noise.

Figures (24)

• Figure 1: A sketch of the fully-connected NN used in this work.
• Figure 2: The workflow of the proposed machine learning-enhanced aerodynamic forces prediction based on sparse pressure sensor inputs.
• Figure 3: A sketch of the numerical simulation of the 2D NACA0015 airfoil.
• Figure 4: Computational mesh and URANS results with $f = \qty{3.979}{Hz}$ at $t = \qty{0.3}{s}$.
• Figure 5: Sensor locations on the airfoil selected by the DEIM based on the experimental data.