Flying a drone takes practice. Most landings involved getting close the ground and then dropping the last inch or two. This is fine for smaller drones, but for on-demand passenger service, landings must be smooth and controlled.
A team at Caltech's Center for Autonomous Systems and Technologies (CAST) is developing a deep neural network (artificial intelligence or AI) that will enable autonomous drones to “learn” how to land safely, quickly, and smoothly. The system they have created, dubbed the "Neural Lander," “learns” through repeated landings, how to modify the approach path and rotor speed to complete a smooth landing.
According to Soon-Jo Chung, "This project has the potential to help drones fly more smoothly and safely, especially in the presence of unpredictable wind gusts, and eat up less battery power as drones can land more quickly." Chung is a Bren Professor of Aerospace in the Division of Engineering and Applied Science (EAS) and research scientist at JPL (Jet Propulsion Laboratory), that Caltech manages for NASA.