Adaptive backstepping PWM stabilization of DC-to-DC converters towards minimum or non-minimum phase equilibria

The linear parameterization assumption often made in adaptive regulation schemes of nonlinear systems can be regarded not only as a counterintuitive assumption but also one with little physical significance. In this paper, a first order Taylor series approximation is proposed that linearizes the natural nonlinear dependencies upon the components of the vector of unknown parameters. This approach is shown to lead to an improved, and more realistic, non-overparameterized backstepping design based on input-output considerations. The results are applied to the direct adaptive output stabilization towards minimum or non-minimum phase equilibria of average pulse-width-modulation controlled DC-to-DC power converters of the Boost type