Drones learn to land on their own using 'fuzzy logic'

Scientists, including one of Indian origin, are using artificial intelligence called fuzzy logic to get drones to navigate and land themselves on moving platforms without any help.

Researchers from University of Cincinnati in the US applied a concept called fuzzy logic, the kind of reasoning people employ subconsciously every day.

Instead of seeing the world in black and white, fuzzy logic allows for nuance or degrees of truth.

"It's the only realistic way that drones will have commercially viable uses such as delivering that roll of toilet paper to customers," said Manish Kumar, associate professor at the University of Cincinnati in the US.

Fuzzy logic helps the drone make good navigational decisions amid a sea of statistical noise. It is called "genetic-fuzzy" because the system evolves over time and continuously discards the lesser solutions, Kumar said.

The researchers successfully employed fuzzy logic in a simulation to show it is an ideal system for navigating under dynamic conditions and are even putting fuzzy logic to the test in experiments to land quadcopters on robots mounted with landing pads at the university lab.

The problem of drones having difficulty in navigating their ever-changing airspace is compounded when the drone tries to land on a moving platform such as a delivery van or even a US Navy warship pitching in high seas, researchers explained.

"It has to land within a designated area with a small margin of error.

"Landing a drone on a moving platform is a very difficult problem scientifically and from an engineering perspective," Kumar added.