Automatic landing on unprepared zone

Abstract

Landing is arguably the most complex and most dangerous phase of flight for an aircraft. As a consequence, automatic landing on unprepared zone is a fundamental function and a key challenge for unmanned aerial vehicles. Many obstacles such as a pylon or any object on the helipad can hinder the descent. Cameras and lidars can be disturbed or deceived by a smoke cloud, low luminosity or obstacles with low reflectance. Therefore, using a combination of different types of sensors for the landing zone clearance diagnosis improves the robustness of the system. Although trajectory generation and optimal control are already widely spread, fewer methods deal with Automatic Decision Making. At Airbus Helicopters we have designed a demonstrator to prove our ability to enable a drone to make its own decision. A drone equipped with an autopilot will be given an order to find the closest helipad and land on it only if it is safe. Otherwise, the drone will fly to the next helipad.