Augmenting the helicopter-ship dynamic interface using an autonomous safe-landing prediction scheme

Abstract

The operational interface between a helicopter and a ship deck is a complex, dynamic and hazardous environment that presents a unique set of challenges to both engineers and pilots alike. Unmanned helicopters require a specific set of conditions to be met and maintained in order to commit to a landing, where large amounts of energy or large angular deviations of the ship deck can cause undesirable and dangerous effects, such as dynamic rollover. An analysis of the prediction problem is given, based on datasets collected from a 70m ship in South African waters, as well as an illustration of a real-time technique based on standard ship measurements that can be used to indicate potential landing periods. This is intended for application as a subsystem that can be used to facilitate autonomous landings of Unmanned Aerial Vehicles (UAVs) while at sea. The system also has potential as a bridge landing aid to allow operators to select ideal landing periods for helicopter operations. Simulations, based on real datasets, illustrate adequate prediction quality ranging from 3 seconds to 20 seconds in the future, depending on the forward prediction coherency for the dataset.