Output-feedback stabilization of nonlinear non-minimum phase systems using neural network

This paper presents an adaptive output-feedback stabilization method for non-affine nonlinear non-minimum phase systems using neural networks. The proposed controller is comprised of a linear, a neuro-adaptive, and an adaptive robustifying control term. The learning rules for adaptive gains, including weights of the neural network, are derived using the Lyapunovpsilas direct method. These adaptation laws employ a suitable output of a linear observer of system dynamics that is realizable. The effectiveness of the proposed scheme will be shown in simulations for the benchmark translation oscillator rotational actuator (TORA) problem.