The Department of the Air Force and the Massachusetts Institute of Technology have jointly demonstrated real-time magnetic navigation on a C-17A Globemaster III using artificial intelligence and machine learning capabilities from the DAF-MIT AI Accelerator.
The MagNav trial was conducted during exercise Golden Phoenix, held from May 11 to 15 at Travis Air Force Base in California, the U.S. Air Force said Friday.
During the demonstration, the MagNav team used AI and machine learning from the AIA’s calibration and positioning neural network and learning from AI models based on data generated from the C-17 to speed up the neural network training process.
The team also worked with global partners through its Magnetic Navigation Open Challenge to enhance the neural network architecture, which eliminates magnetic noise produced by the aircraft.
Personnel from the Air Force Research Laboratory Sensors Directorate, Air Force Institute of Technology Autonomy and Navigation Center, MIT and MIT Lincoln Laboratory supported the demonstration.
SandboxAQ also secured approval to deploy a quantum magnetometer ahead of its schedule under the Small Business Innovation Research program.
