Demonstration of a model-free backstepping control on a 2-DOF laboratory helicopter

A model-free backstepping (MFBS) control design method is proposed to deal with problems of model uncertainties, dynamic coupling and time-varying coefficients of nonlinear dynamic systems. A popular benchmark system of 2-DOF laboratory helicopter is used to demonstrate and validate the theoretical development. The MFBS method makes use of techniques including the normal form, model estimation and cancellation, and control coefficient substitution. The model estimation is based on the input and output data only. The model free backstepping control is designed with the Lyapunov method and its stability is proven. The equivalence of the control designed with the control coefficient substitution to that with the full knowledge of the model is discussed. Simulation and experimental results are presented to show the effectiveness of the MFBS control design in comparison with a common LQR control.