ETHZ ASL

Logo

Datasets and open-source code from the Autonomous Systems Lab (ASL) at ETH Zurich, led by Prof. Roland Siegwart.

In-Wing Pressure Measurements for Airspeed and Airflow Angle Estimation

Year: 2022
DOI: 10.3929/ethz-c-000788334

Hardware designs and datasets for in-wing differential pressure sensing on a small UAV (EasyGlider 4), covering wind tunnel and flight tests for airspeed and airflow angle estimation and deep stall/high angle-of-attack maneuvers. Preview imagery and PCB/STEP files to be rehosted with the dataset.

Dataset Description

  • HDF5 logs for raw and calibrated data, each with wind tunnel and flight datasets
  • Includes PX4-based navigation data (ulog), synchronized at 40 Hz with preliminary airspeed estimates
  • Calibrated datasets provide airspeed, airflow angles, and current following the paper’s procedure
  • Key fields include differential pressures, airspeed, angles of attack/sideslip, vehicle attitude, and PX4 state
  • Example Python script for reading/plotting; MATLAB HDF5 tools also applicable

Hardware Designs

  • Differential pressure module STEP files
  • PCB designs as KiCAD projects

Downloads

Citation

G. A. Heinrich, S. Vogt, N. R. Lawrance, T. J. Stastny, R. Y. Siegwart,
“In-Wing Pressure Measurements for Airspeed and Airflow Angle Estimation and High Angle-of-Attack Flight”, Journal of Guidance, Control, and Dynamics, 2022. DOI: 10.2514/1.G006412.

Tags: uav, aerodynamics, sensing, pressure, flight