Towards Automation of Helicopter Landings on Ship Decks Using Integer Programming and Model Predictive Control

A novel method for the automated control of helicopters in landing maneuvers on ship decks is proposed, which combines integer programming and model predictive control (MPC). The helicopter is first brought sufficiently close to the landing deck using a standard MPC. In the final phase of the mission, termed the rendezvous phase, implementation of the novel control design method allows the MPC to rapidly adapt to the landing deck state via a variable prediction horizon. The control design problem includes an integer variable vector which is used to frequently estimate an appropriate prediction horizon for the MPC during the maneuver. This process mimics the manner in which human pilots act, by repeatedly estimating how long ahead during the ship landing maneuver their actions are effective to safely accomplish the mission. The performance of the proposed procedure is evaluated via numerical simulations using a nonlinear helicopter-ship dynamics interface that captures significant characteristics of the helicopter and ship dynamics, as well as the ship airwake effects. These simulations show that the proposed method can be effective in satisfying rendezvous conditions for helicopter ship landing even when the moving deck has a very high energy index.