Aerospace Sciences

Machine learning, flight experiments highlight progress in aerospace environments


The Atmospheric and Space Environments Technical Committee encourages the exchange of information about the interactions between aerospace systems and their surroundings.

In May, UBIQ Aerospace of Norway announced that it completed testing of its holistic ice protection system, D•ICE, consisting of an ice detection sensor, a heated air speed sensor, heated wings and a heated propeller. The system was designed for small fixed-wing drones and has several modules that protect all icing-exposed aircraft components, enabling operations and sustained flight in potential and known icing conditions. UBIQ Aerospace flight tested all system components in freezing conditions of minus 13 degrees Celsius with light wind. Tests were performed with a Maritime Robotics PX-31 Falk in collaboration with a defense project for the Norwegian Armed Forces and the Norwegian Defense Research Establishment.

In March, NASA completed an icing wind tunnel test campaign with a subscale propeller for urban air mobility applications. In-flight icing is an important safety consideration for UAM vehicles, but there are limited experimental data and analysis tools that can be used to demonstrate safe operation. The test campaign was conducted in the Icing Research Tunnel at NASA’s Glenn Research Center in Ohio to gather data on three geometrically scaled propellers over a range of icing conditions. Acquired data included 3D scans of ice shapes, high-speed video of ice shedding events and changes to motor power requirements during icing exposure. Researchers will use the test results for the continued development and validation on icing analysis tools for UAM applications. The completion of this test campaign marks the first time that NASA has developed and tested a rotating test article in an icing wind tunnel without proprietary industry hardware, ensuring that all results can be made available to the public.

A team from NASA and the Johns Hopkins University Applied Physics Laboratory in Maryland reported in a June press release that the Parker Solar Probe “has flown close enough to the sun to detect the fine structure of the solar wind close to where it is generated at the sun’s surface.” Most of the space around the sun is filled by these coronal holes, which are created when magnetic field lines “emerge from the surface without looping back inward, thus forming open field lines that expand outward,” the press release reads. Based on the team’s analysis, “the coronal holes are like showerheads, with roughly evenly spaced jets emerging from bright spots where magnetic field lines funnel into and out of the surface of the Sun. When oppositely directed magnetic fields pass one another in these funnels, which can be 18,000 miles across, the fields often break and reconnect, slinging charged particles out of the Sun.” The Parker Solar Probe orbits the sun with the closest approach distance of about 10 solar radii — the closest any human object has every managed to repeatedly visit.

NASA continues to study changes in the near-Earth space environment. In August, NASA selected four Space Weather Centers of Excellence to provide data on current conditions as well as forecasts to give space and ground systems time to prepare for space weather events: the Space Weather Research and Technology Applications Center of Excellence at Boston College; Space Weather Operational Readiness Development Center at the University of Colorado, Boulder; Center for All-Clear Solar Energetic Particle Forecast at the University of Michigan, Ann Arbor; and Center of Excellence for Advanced Forecasting of Drag for Enhanced, Sustainable, and Conscientious Space Operations at the University of West Virginia, Morgantown. The Space Weather Centers of Excellence program takes advantage of lessons learned from ongoing and past science centers and promotes synergistic, coordinated efforts to transform space weather capabilities and preparedness.

Contributors: Richard Hann and Paul von Hardenburg

Machine learning, flight experiments highlight progress in aerospace environments