Model predictive control for autonomous unmanned helicopters

Purpose - The purpose of this paper is to investigate the feasibility of controlling a small-scale helicopter by using the model predictive control (MPC) method. Design/methodology/approach - The MPC control synthesis is employed by considering five linear models representing the flight of a small-scale helicopter from hover to high-speed cruise. The internal model principle is employed for the trajectory tracking design. Findings - It is found that the MPC handles well the transition problems between the models, yields satisfactory tracking control performance and produces a suitable control signal. The performance of the tracking control of the helicopter is considerably influenced by the parameter selection in the states and inputs weighting matrices of the MPC. Simulation results also showed that faster dynamics, coupling problems, input and output constraints and changing linearized multi-inputs multi-outputs dynamics models in the small-scale helicopter can be handled simultaneously by the MPC controller. Research limitations/implications - The present study is limited for the application of MPC for the control of small-scale helicopters with non-aggressive maneuvers. Practical implications - The result can be extended to design a full envelope controller for an autonomous small-scale helicopter without the need to resort to a conventional gain scheduling technique. Originality/value - Helicopter control system designs using MPC with a single either linear or non-linear model have been studied and reported in numerous literatures. The main contribution of the paper is in the application of MPC to handle the control problems of a small-scale helicopter defined as a mathematical model with several different modes during a flight mission.