Application of linear and nonlinear model predictive control for reducing helicopter cross-couplings

Abstract

Single rotor helicopters have strong cross couplings manifested in the response to control between longitudinal and lateral-directional commands. These couplings cannot be eliminated by static considerations of control phasing since the cross coupling is a function of the frequency content of the control input. This paper proposes a model predictive control (MPC) method to alleviate the strong helicopter cross-couplings (e.g., roll due to pitch control and pitch and roll due to collective control). The advantage of using MPC is that this well-known discrete method in optimal control has the powerful capability of inclusion of look-ahead information about the frequency content of the input and states constraints. The paper investigates both whether linear and nonlinear MPC are suitable for online application to helicopters cross-couplings reduction. It is demonstrated that both linear and nonlinear MPC are effective methods in reducing cross-coupling effects on a Bolkow Bo-105 helicopter even when external disturbances and model errors are present. Using the ADS-33 handling qualities (HQs) criteria on crosscouplings, the paper demonstrates that Level 2 and 3 HQs of the uncontrolled helicopter are improving to Level 1 HQs once the MPC controller is switched on with almost ideal off-axis rate response corresponding to no couplings. Furthermore, compared to a classical proportional-integral-derivative (PID) controller, the MPC performs 55 percent to 80 percent better than the PID with and without simulation model uncertainty or disturbance introduced in the controller.