A Barrier Function Based Feedback Linearization Method for On-line Output Tracking Control of Non- minimum Phase Systems

This paper presents a new method for solving the output tracking problem in nonlinear non-minimum phase systems using feedback linearization method. The proposed method converts the problem to an equivalent optimization problem where the constraints of which include minimum phase conditions and the cost function is the Euclidean distance between the system output and the parameterized redefined output. To solve this constrained optimization problem, it is first approximated to an unconstrained one using the barrier method and then, through the steepest descent algorithm, optimization parameters (the redefined output parameters) are iteratively obtained in an on-line manner such that at each time the minimum phase conditions are satisfied. Indeed, solving the optimization problem leads to find a minimum phase output as close as possible to the system output. On the other hand, since this minimum phase output tracks the reference signal using feedback linearization method, the system output converges to the reference signal.