GPS can be great for navigating your local streets, but when it comes to navigating military planes and other aircraft, it is very prone to error.
The Air Force is now on a mission to use the Eath’s magnetic field as an alternative to GPS, according to Defense One.
The idea has been in the works for a few years with Aaron Canciani, assistant professor of electrical engineering at the Air Force Institute of Technology, who tested the concept in 2017.
Canciani tested to see if any magnetic sensors on an airplane could measure the intensity of Earth’s magnetic fields.
If so, these sensors would be able to locate the aircraft based on its location relative to the magnetic fields of our planet. His experience has shown that this method is a viable option as an alternative to GPS.
But just because the method can be viable doesn’t make it seem easy or straightforward. The electrical operations of the aircraft itself, interfere with a sensor’s ability to detect field strength.
This is where the Air Force decided to use artificial intelligence. AI is famous for canceling noise from sensor readings. This results in a better, more precise signal.
Researchers from the Air Force and MIT’s artificial intelligence accelerator community, have teamed up to work on noise cancellation. They published their own article in July.
They found that magnetic field readings can be accurate to ten meters. Compared to GPS readings, which are accurate up to three meters, this can seem like a real drawback. But magnetometer readings have a key advantage: they’re much more difficult to block.
Because of the size of the earth and the magnetic field, it takes a lot to scramble a signal from the earth, and by many, it means on the scale of a nuclear explosion.
Other than that, it would take a giant scale of a machine to block what comes from the earth’s crust. But you can also override it with machine learning.
The Air Force, working with MIT, is now looking to find better tools for cleaning magnetic field readings. They have released a new joint accelerator program that ends on August 28.