Shrinking Horizon Model Predictive Control Method for Helicopter–Ship Touchdown

Abstract

A model predictive control (MPC) framework with a fixed maneuver horizon and shrinking prediction and control horizons is presented that, at each time step, minimizes the most accurate prediction of a complete cost for a discrete linear system, subject to constraints. Methods of weight selection to ensure strong convexity of the cost, which makes the quadratic programming problem associated with MPC numerically more tractable, are discussed. A continuous-time flexible-blade helicopter dynamic model is discretized, and the resulting model is used to demonstrate this control design method in ship landing and touchdown maneuvers. Inequality constraints, ship-induced turbulence, and parametric uncertainty are gradually included in the design and analysis. Several case studies are used to illustrate the effectiveness of this control method in landings on ships that experience quiescent and nonquiescent motions.