摘要:AbstractIn this paper, a decentralised tracking control based on extended linearisation techniques is presented for a two rotor laboratory helicopter with two degrees of freedom. Employing Lagrange's equations, a nonlinear control-oriented state-space model can be derived. It is rewritten in a quasi-linear form – without any simplifications – with state-dependent matrices as a basis for the decentralised feedforward and feedback control design. The control task consists in tracking accurately desired trajectories for both the azimuth angle and the pitch angle. Due to unmeasurable states as well as uncertainties stemming from both model simplifications at modelling and unknown disturbance torques, an unscented Kalman filter is employed and combined with a discrete-time implementation of the nonlinear control law. The eficiency of the proposed controller is demonstrated by results from an experimental set-up.
关键词:KeywordsMultivariable systemsextended linearisationunscented Kalman filterhelicopter control